Skip to main content

Full text of "Phosphorus forms and retention in a sandy soil receiving dairy waste effluent"

See other formats


PHOSPHORUS FORMS AND RETENTION IN A SANDY SOIL RECEIVING DAIRY 

WASTE EFFLUENT 



By 

ABDULLAH AL-SHANKITI 



A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL 

OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULHLLMENT 

OF THE REQUIREMENTS FOR THE DEGREE OF 

DOCTOR OF PHILOSOPHY 

UNIVERSITY OF FLORIDA 

2000 



ACKNOWLEDGMENTS 
Sincere appreciation and gratitude go to my major advisor. Dr. D. A. Graetz, for 
giving me all the help I needed throughout the course of my study. I would also like to 
thank my committee members-Dr. W. G. Harris, Dr. R. D. Rhue, and Dr. R. Nordstedt-- 
for their guidance, comments, and suggestions. I am grateful to Dr. V. Nair for her 
insightful comments and suggestions, to Dr. K. R. Woodard for his help and support, and 
to D. Lucas for her encouragement and support. 



u 



; - ? ; . TABLE OF CONTENTS 

page 

ACKNOWLEDGMENTS iii 

LIST OF TABLES vi 

LIST OF FIGURES viii 

ABSTRACT x 

CHAPTERS 

1 INTRODUCTION 1 

Statement of the Problem 3 

Objectives 4 

Review of Literature 5 

Soil Phosphorus 6 

Phosphorus Accumulation 8 

Phosphorus Forms and Fractionation 1 1 

Phosphorus Retention 13 

Downward P movement 20 

Manure Management 22 

Dissertation Format 23 



2 PHOSPHORUS ACCUMULATION IN A SANDY SOIL RECEIVING DAIRY 
WASTE EFFLUENT 25 

Introduction 25 

Materials and Methods 27 

Experiment Location and Design 27 

Soil Selection and Sampling 27 

Soil Characterization 27 

Effluent Application and Characterization 28 

Statistical Analysis 28 

Results and Discussion 30 

Soil Properties Prior to Effluent Application 30 

Effect of Application Rate and Cropping Systems 33 

Summary and Conclusions 42 



ui 



3 PHOSPHORUS FORMS AND FRACTIONATION IN A SANDY SOIL 
RECEIVING DAIRY WASTE EFFLUENT 44 

Introduction 44 

Materials and Methods 46 

Experiment Location and Design 46 

Soil Selection and Sampling 46 

Fractionation Scheme 47 

Statistical Analysis 48 

Results and Discussion 48 

Study Site 54 

Summary and Conclusions 61 



4 PHOSPHORUS RETENTION IN A SANDY SOIL RECEIVING DAIRY WASTE 
EFFLUENT 63 

Introduction • 63 

Materials and Methods 65 

Experiment Location and Design 65 

Soil Selection and Sampling 65 

Soil Characterization 65 

Calculations 67 

Statistical Analysis 68 

Results and Discussion 68 

Relative Phosphorus Adsorption (RPA) 68 

Degree of Phosphorus Saturation (DPS) 71 

Langmuir Adsorption Parameters 73 

Summary and Conclusions 77 



5 DOWNWARD PHOSPHORUS MOVEMENT ASSESSMENT IN A SANDY SOIL 
RECEIVING DAIRY WASTE EFFLUENT 81 

Introduction 81 

Materials and Methods 83 

Experunent Location and Design 83 

Soil Selection and Sampling 83 

Soil Characterization 84 

Statistical Analysis 84 

Results and Discussion 85 

Summary and Conclusions 95 



6 UTILIZATION OF DAIRY WASTE EFFLUENT THROUGH SEQUENTIAL 
CROPPING 97 



IV 



Introduction 97 

Materials and Methods 98 

Experiment Location and Design 98 

Sampling and Analysis 99 

Results and discussion 100 

Summary and Conclusions 105 

7 SUMMARY AND CONCLUSIONS 108 

APPENDIX • ' ^ ^-. 

SELECTION OF SOIL: SOLUTION RATIO 114 

LIST OF REFERENCES 118 

BIOGRAPfflCAL SKETCH 126 



A ■ 



LIST OF TABLES 



Table Page 

Table 2-1. Average annual concentrations (mg/L) of ammonium nitrogen (NH4 -N), total 
Kjeldahl nitrogen (TKN), soluble reactive P (SRP), and total P (TP) in 
effluent applied to the study site. Numbers in parentheses are standard 
deviations 28 

Table 2-2. Selected characteristics of typical Kershaw sand (Soil Survey Staff, Gilchrist 

County, Florida, 1973) compared to the study site 31 

Table 2-3. Mehlich I-extractable elements concentrations and total P (TP) in "native" soil 
(n = 3 profiles) and study site (n ^ 12 profiles) soil profiles prior to beginning 
of the study 32 

Table 2-4. Statistical evaluation of TP data for the three-year study period 34 

Table 2-5. Regression equation relating Mehlich I-P to the independent variables Mehlich 

I-Ca, Mg, and Fe. (n=432) 42 

Table 3-1 . P values (mg/kg) in each fi^action within a soil depth increment at the 

begiimmg (1996) and end of the study period (1998) (n = 12 profiles). Values 
are Least Square Means (LSM) 49 

Table 3-2. Percentage of P in each fraction within a soil depth increment at the begiiming 
(1996) and end of the study period (1998)(n = 12 profiles). Values are Least 
Square Means (LSM) 50 

Table 3-3. Increases in each fraction within a soil depth increment between the beginning 

(1996) and end ofthe study period (1998) 52 

Table 3-4. Mean concentration of Mehlich I extractable elements (mg/kg) in the soil 
profile ofthe study site in 1996 prior to the application of effluent (n = 12 
profiles) 54 

Table 3-5. P values (mg/kg) in each ofthe fractions within a soil depth increment at the 

native site (n = 3 profiles). Values are Least Square Means (LSM) 55 



VI 



Table 3-6. Percentage of P in each of the fractions within a soil depth increment at the 

native site (n = 3 profiles). Values are Least Square Means (LSM) 55 

Table 4-1 . RPA values within the soil profile of the study site (n = 12 profiles) prior and 
after to application of effluent compared to the "native soil" (n = 1 profile). 
Values are Least Square Mean (LSM) 69 

Table 4-2. Multiple regression equations relating RPA to a) Mehlich I (DA) Al, Fe and P, 

b) Oxalate AI, Fe, and P in 1996 (prior to the application of effluent) (n = 72). 72 

Table 4-3. DPS - 1^ % values within the soil profile of the study site (n = 12 profiles) 
prior and after to application of effluent compared to the "native" soil (n == 1 
profile). Values are Least Square Means (LSM) 72 

Table 4-4. DPS - 2^ % values within the soil profile of the study site (n = 12 profiles) 
prior and after to application of effluent compared to the "native" soil (n = 1 
profile). Values are Least Square Means (LSM) 74 

Table 4-5. Comparison of Langmuir parameters (Smax, EPCo, k) and So mean values of 
different horizons within the soil profile prior to the application of effluent in 
1996 and after two years of effluent application in 1998 76 

Table 5-2. Mehlich I-extractable elements concentrations and total P (TP) in "native" soil 
(n = 1 profile) and study site soil profiles (n = 12 profiles) prior to the start of 
the study 87 

Table 5-3. Changes in WSP concentration within the soil profile under high application 
rate after the application of effluent ( 1 998) vs. prior to the application of 
effluent (1996) 90 

Table. 5-4. Changes in WSP concentration within the soil profile under the low 
application rate after the application of effluent (1998) vs. prior to the 
application of effluent (1996) 90 

Table 6-1. P removed (kg/ha) by the com- forage sorghum-rye cropping system under 

high and low application rates during the 1996-97 and 1997-98 seasons. (Data 
obtained from Woodard et al. 2000) 101 

Table 6-2. P removed (kg/ha) by the perennial peanut-rye cropping system under high 
and low application rates during the 1996-97 and 1997-98 seasons. (Data 
obtained from Woodard et al. 2000) 103 

Table 6-3. Average dry matter yield of the com- forage sorghum-rye during the 1996-97 

and 1997-98 seasons 106 

Table 6-4. Average dry matter yield of the perennial peanut-rye during the 1996-97 and 

1997-98 seasons 106 



vu 



LIST OF FIGURES 



Figure ' 2i£§ 

Figure 2-1 . Average total P (TP) concentrations in the soil profile under the high rate 
appUcation prior to application of effluent (1996) and after effluent 
application (1997 and 1998). Values are LSM± Std. Error 36 

Figure 2-2. Average total P (TP) concentrations in the soil profile under the low^ rate 
application prior to application of eifluent (1996) and after effluent 
application (1997 and 1998). Values are LSM± Std. Error 37 

Figure 2-3. Mehlich I-extractable P concentrations in the soil profile prior to start of the 
study and after two years of effluent application (1998). Values are LSM± 
Std. Error 38 

Figure 2-4. Mehlich I-extractable P concentrations for cropping systems under the high 

rate effluent application in 1998. Values are LSM± Std. Error 40 

Figure 2-5. Mehlich I-extractable P concentrations for cropping systems under the low 

rate effluent application in 1998. Values are LSM± Std. Error 41 

Figure 3-1. Al-Fe-associated P (mg/kg) within the soil profile at the begiiming 1996 and 

end of the study period (1998). Values are LSM± Std. Error 51 

Figure 3-2. Labile P values (mg/kg) within the soil profile at the beginning (1996) and 

end of the study period (1998). Values are LSM± Std. Error 57 

Figure 3-3. Ca-Mg associated P values (mg/kg) within the soil profile at the beginning 

(1996) and end of the study period (1998). Values are LSM± Std. Error 58 

Figure 3-4. Residual-P values (mg/kg) within the soil profile at the begiiming (1996) and 

end of the study period (1998). Values are LSM± Std. Error 60 

Figure 4-1. Relationship between Degree of P saturation (DPS - 1) calculated fi-om 

oxalate extractable-P and Degree of P Saturation calculated fi-om Mehlich I 
(DPS - 2) for soil samples fi-om the study site 75 



vni 



Figure 4-2. Relationship between Degree of P saturation calculated from oxalate 
extractable-P (DPS - 1) and equilibrium P concentration (EPCo) for soil 
samples from the study site 78 

Figure 4-3. Relationship between Degree of P saturation calculated from oxalate 

extractable-P (DPS - 1) and soluble P (Po) mg/L for soil samples from the 
study site 79 

Figure 5-1. Mean water soluble P (WSP) concentrations within the soil profile of the 

study site under the high rate effluent appUcation prior to the application of 
effluent in 1996 and after effluent application in 1998. Values are LSM± Std. 
Error 88 

Figure 5-2. Mean water soluble P (WSP) concentrations within the soil profile of the 
study site under the low rate effluent application prior to the application of 
effluent in 1996 and after effluent application in 1998. Values are LSM± Std. 
Error 89 

Figure 5-3. Mean water soluble P (WSP) concentration within the soil profile of the study 
site prior to the application of effluent in 1996 and after effluent application in 
1998. Values are LSM± Std. Error 92 

Figure 5-4. Labile-P concentration within the soil profile of the study site prior to the 

application of effluent in 1996 and after effluent application in 1998. Values 
are LSM± Std. Error 93 

Figure 6-1. P removal (kg/ha) of corn-forage sorghum-rye during the 1996-97 and 1997- 

98 seasons 102 

Figure 6-2. P removal (kg/ha) of perennial peanut-rye during the 1996-97 and 1997-98 

seasons 104 



IX 



Abstract of Dissertation Presented to the Graduate School 

of the University of Florida in Partial Fulfilhnent of the 

Requirements for the Degree of Doctor of Philosophy 

PHOSPHORUS FORMS AND RETENTION IN A SANDY SOIL RECEIVING DAIRY 

WASTE EFFLUENT 

By 

Abdullah Alshankiti 

May 2000 
Chairman: D.A. Graetz 
Major Department: Soil and Water Science 

Currently there are major concerns about the potential negative effects of nutrient 
losses from the waste of dairy farms on surface and ground water quaUty. In many 
confined livestock production systems, manures are normally applied at a rate designed 
to meet crop N requirements. However, this often results in a buildup of soil P above 
amounts required for optimal crop yield and increases the chances for P losses from 
source areas to water bodies. 

This research, conducted at a dairy farm in north Florida, investigates the status of 
soil P under two main treatments of dairy waste effluent and two cropping systems. The 
N application rates were 448 and 896 kg/ha/yr which correspond to P loading of 1 12 and 
224 kg/ha/yr. The cropping systems were perennial peanut-rye (P-R) and com- forage 
sorghum-rye (C-FS-R). The objectives were to: (1) examine the accumulation of P in the 
soil profile, (2) quantify and characterize P forms m the soil profile, (3) quantify and 



characterize P retention in the soil profile, (4) determine P uptake by the cropping 
systems, and (5) assess the downward movement of P. 

The study site, mapped as Kershaw sand, appears to have been heavily loaded 
with animal waste (47 mg/kg Mehlich I-extractable P (MI-P) in the native area vs. 283 
mg/kg in the study site surface soils). The MI-P increased significantly with high effluent 
rate application, particularly under the P-R cropping system, which suggests that the C- 
FS-R cropping system may be more effective in P removal than the P-R cropping system. 
Total P (TP) increased from 343 mg/kg in 1996 to 689 mg/kg in 1998. Water soluble P 
(WSP) increased but primarily m the lower depths of the soil profile under both 
treatments. 

Al- and Fe-associated P constituted the major proportion (up to 60%) of the TP in 
the soil profile. Labile-P accoimted for 1 8 to 30%, and Ca- and Mg-associated P 
accounted for about 10% of TP. Water soluble- and labile-P concentrations from 1996 
and 1998 indicated a downward movement of P in the soil profile. These same data 
coincided with a decrease in retention capacity as determined by "Relative Phosphorus 
Adsorption" (RPA). Degree of P Saturation (DPS) data indicated that the surface horizon 
is more likely to release P than the deeper depths. The conclusions drawn from DPS were 
in agreement with the conclusions arrived at from the soil adsorption capacity and 
equilibrium phosphorus concentration (EPCo). 

Phosphorus removal was higher for the C-FS-R than for the P-R cropping system. 
The removal values agreed with published P uptake for such crops, but crop uptake did 
not alter the high level of soil P that was already present before application. When soil 
test P levels in the soil exceed optimum values for crop production, the application of 



XI 



dairy waste based on estimated N requirement may not be appropriate on heavily P 
loaded sandy soil such as the soil at the study site. 



xu 



CHAPTER 1 
INTRODUCTION 

The impact of current agriculture management practices in farmland or animal- 
related activity on water quality is well documented. Runoff from agricultural land is one 
of the major sources of nonpoint-source pollution. The USEPA has identified agriculture 
nonpoint-source pollution as the major source of stream and lake contamination that 
prevents attainment of water quality goals identified in the Clean Water Act (Parry, 1998; 
USEPA, 1996). The transport of phosphorus (P) to surface water can lead to accelerated 
eutrophication of these waters, which limit their use for fisheries, recreation, industry, or 
drinking. Although nitrogen (N) and carbon (C) are also associated with accelerated 
eutrophication, most attention has focused on P because P often limits eutrophication and 
its control is of prime importance in reducing the accelerated eutrophication of surface 
water (Thomann and Mueller, 1987) 

Most P in agriculture soils is found either as insoluble precipitates of Ca, Fe and/ 
or Al or as a constituent of a wide range of organic compounds. Water moving across or 
through soils removes both soluble P and sediments enriched with P, usually with the 
lighter, fine sized particles such as clays and organic matter. The soluble or particulate P 
then either can enter a flowing water body where it can be deposited as sediment or can 
be carried directly into a lake or pond. Phosphorus can also leach downward in the soil, 
perhaps to a tile drainage system or to ground water, where subsurface transport can then 
discharge the P into a stream or lake (Sharpley and Halvorson, 1994). 



Most of the P that enters aquatic ecosystems comes from agricultural use. Phosphorus is 
added to lands as fertilizers, organic solids, wastewater, and feeds. It is estimated that 
42,660 Mg of fertilizers was used during 1996 in Florida (Reddy et al., 1999). Fertilizer P 
is primarily in inorganic form, which is bioavailabile and can be a major source of P for 
many ecosystems. For example, fertilizer P accounted for 5 1% of P imports to the 
Okeechobee Basin (Boggess et al., 1995). Another significant source of P input to the 
lake was the dairy farming and beef cattle ranching north of the lake which accounted for 
about 49% of the TP input to the lake (Federico et al., 1981). Thus, optimal dairy waste 
management practices are more necessary than ever; more cows on limited land area 
increase the likelihood of environmental problems resulting from mismanagement of 
dairy farm wastes. A dairy waste management system should account for the fate of 
nutrients that may be of environmental concern. The overall goal of sound agronomic 
and environmental management programs for soil P is to maximize plant growth, while 
minimizing losses of P to surface waters (Lanyon, 1994). It is important, therefore, to 
understand the role of soil reactions in controlling the availability of soil P for plant 
uptake or loss in erosion, surface runoff, and leaching. Amounts of P exported from 
watersheds are tied to watershed hydrology, soil P content, and amount of P added as 
fertilizer or manure. This assumes in most cases that P export from watersheds occurs in 
surface rather than subsurface runoff, although it is recognized that in some regions of the 
US dominated by sandy or organic soils P can be transported in subsurface drainage 
waters. Generally, the P concentration in water moving through the soil profile is small 
due to sorption of P, except in acid organic or peaty soils where the adsorpfion affinity 
and capacity for P retention are low (Sims et al., 1998). Similarly, sandy soils with low P 



sorption capacities, waterlogged soils, and soils with preferential flow through 
macropores and earthworm holes are susceptible to P movement (Sharpley and Syers, 
1979). ■ 

Statement of the Problem 

In 1990, the Middle Suwannee River area was approved as a Hydrologic Unit 
Area project based on data generated by the Florida Department of Environmental 
Protection. These data showed an elevated concentration of nitrate-nitrogen in the 
Floridan Aquifer in the Suwannee River Basin, especially in areas of intensive 
agricultural activity. Phosphorus concentrations in the Suwannee River ranged from 0.40 
to 0.49 mg/L which were 6.4 times the median regional value of north Florida streams. 
The Hydrologic Unit Area program was developed to reduce or prevent water quality 
degradation of the Floridan Aquifer and the Suwannee River resulting from agricultural 
operations. Management of nutrients (potential contaminants) in dairy waste effluent 
through spray field crop production systems is an important component in the overall 
scheme for protecting ground and surface water from elevated levels of N and P. The use 
of inappropriate crop management technology under a dairy effluent irrigation system 
can lead to the loss of N to the ground water. Uptake of nutrients by agronomic crops 
sequenced over time is an effective, economical, and environmentally sound means of 
nutrient recovery. Cropping systems designs are needed to meet environmental demands 
by maximizing nutrient uptake while meeting the needs of dairy producers. 

The Use of Dai ry Manure Effluent in A Rhizoma (Perennial) Peanut Based 
Cropping Systems for Nutrient Recovery and Water Oualitv Enhancement is a research 
project established under the Hydrologic Unit Area project (HUA). The objective of this 



project was to evaluate five cropping systems grown under a dairy effluent disposal 
irrigation system, comparing their effectiveness in nutrient recovery and maintenance of 
acceptable levels of N and P in ground water. The cropping systems were corn-forage 
sorghum-rye, com-bermudagrass-rye, bermudagrass-rye, perennial peanut-rye, and corn- 
perennial peanut-rye. The N application rates were 448, 672 and 896 kg/ha/yr which 
correspond to P loadings of 112, 168 and 224 kg/ha/yr. My study was a component of 
this project and addressed P forms and retention in the soil profile under two cropping 
systems (corn-forage sorghum-rye and perennial peanut-rye) and two N application rates 
(448 and 896 kg/ha/yr) which correspond to P loadings of 1 12 and 224 kg/ha/yr. 

In order to achieve the objectives mentioned below, two cropping systems were 
chosen from the main study: corn-forage sorghum-rye and perennial peanut-rye. The 
workload associated with evaluating each treatment in the overall project would have 
been prohibitive, therefore treatments were selected which provide representative data 
with regard to the fate of P in the various cropping systems. The former is commonly 
used by North Florida dairies (Staples, 1997). Recently, perennial peanut has been 
identified as promising for its potential of continuous nutrient recovery over an extended 
period of the year and for production of high quality forage 

Objectives 
The main objective of this research was to study the effect of dairy waste effluent 
application on P accumulation, forms, and retention in the soil profile of a sandy soil 
under two cropping systems. The cropping systems were corn-forage sorghum-rye, which 
represent the traditional crops for the Middle Suwannee River area, and perennial peanut- 



rye, an improved cropping system to be introduce to the area. The specific objectives and 
hypotheses of this research were as follow: 

Objective 1: Quantify and characterize inorganic P forms in the soil profile of the 
chosen cropping systems with increasing effluent P application. 

Hypothesis: Application of dairy waste effluent will increase P levels in the soil 
resulting in an accumulation of P in the soil profile. 

Objective 2: Quantify and characterize P retention capacity in the soil profile. 

Hypothesis: Soil retention capacity will decrease with continuous addition of 
dairy waste effluent and may induce a downward movement of P. 

Objective 3: Determine P uptake by the chosen cropping systems under two rates 
of effluent application. 

Hypothesis: P accumulation in soil profile will decrease with increasing plant 
uptake. 

Review of Literature 
Phosphorus (P) is an integral and essential part of the food production system, but 
P doesn't occur abundantly in most soils. Total P concentration in surface soils varies 
between about 0. 02 and 0.10% (Tisdale et al., 1993). The native P compounds are mostly 
unavailable for plant uptake, some being highly insoluble. When soluble sources of (P) as 
those in fertilizer and manure are added to soils, they are fixed or are changed to 
unavailable forms and in time, react further to become highly insoluble forms. Farmers 
commonly apply more P in fertilizers and manure than is removed by the crops. In time, 
soil P levels increas often to high enough levels to reduce significantly ftiture 



requirements for P fertilizers and cause a buildup of P reserves in the soil profile (Brady, 
1990). - 

Soil Phosphorus 

Phosphorus in agriculture soils is found in inorganic and organic forms. Inorganic 
forms represent 50-70% of soil P, although this fraction can vary from 10 to 90% 
(Pierzynski et al., 1994). Inorganic forms are typically hydrous sesquioxides and 
insoluble precipitates of Ca, Fe and/or Al. Organic P varies between 15 and 80% in most 
soils (Tisdale et aJ., 1993). The quantity of organic P in soil generally increases with 
increasing C and /or N. Many of the organic P compounds in soils have not been 
characterized, but most are esters of orthophosphoric acid and have been identified 
primarily as inositol phosphate, phospholipids, and nucleic acids. Organic P turnover in 
soils is a result of P mineralization and immobilization reactions which, in general, are 
similar to those of N as both processes occur simultaneously in soils. The initial source of 
soil organic P is plant and animal residue, which is degraded through microbial activity to 
produce other organic compounds and release inorganic P (Tisdale et al., 1993). 

There is an interrelationship between the various forms of P in soils. The decrease 
in soil solution P concentration with absorption by plant roots is buffered by both 
inorganic and organic fractions in soil. Primary and secondary P minerals (nonlabile P) 
dissolve to resupply H2P047 HP04^" in solution. Inorganic P adsorbed on mineral and 
clay surfaces as H2P04' or HP04^" (labile inorganic P) also can desorb to buffer P in 
solution P. 

Numerous soil microorganisms digest plant residues containing P and produce 
many organic P compounds in soil. These organic P compounds can be mineralized 
through microbial activity to supply inorganic P. Soil solution P is often called the 



'intensity factor', while the inorganic adsorbed P and organic labile P fractions are 
collectively called the 'quantity factor'. Maintenance of solution P concentration or 
(intensity) for adequate P nutrition in the plant depends on the ability of labile P 
(quantity) to replace soil solution P taken up by the plant. The ratio of quantity to 
intensity is called the 'capacity factor' which expresses the relative ability of the soil to 
buffer changes in soil solution P. Generally, the larger the capacity factor, the greater the 
ability to buffer solution P. The P cycle can be simplified to the following relationship: 

Soil solution <- -¥ Labile P^ -^ nonlabile P 

Labile P is the readily available portion of the quantity factor that exhibits a high 
dissociation rate and rapidly replenishes solution P. Depletion of labile P causes some 
nonlabile P to become labile, but at a slow rate. Thus, the quantity factor comprises both 
labile and nonlabile fraction (Tisdale et al, 1993). 

The division of P in the soil's solid phase into the labile and nonlabile forms 
comes about from a kinetic consideration. From a mechanistic point of view, P in the 
soil's solid phase can be classified by yet another way into adsorbed P and crystalline P. 
The first refers to P adsorbed on active surfaces in the soil, and the second to distinct P 
compounds either formed as reaction products, or inherently present in the soil matrix. 
The two types of categorization (i.e., labile vs. nonlabile and adsorbed vs. crystalline) are 
not synonymous, ahhough a great deal of overlap exists between the two. The labile P 
does not represent a precisely distinct phase of solid phase P, but one that has arbitrary 
boundaries of time and other procedural factors. Any loss of precision in defining labile P 
is paralleled by an equal uncertainty in defining the remaining P (Olsen and Khasawneh, 
1980). Phosphorus amendments, in either organic or inorganic form, are needed to 



8 

maintain adequate available soil P for plant uptake. Once applied, P is either taken up by 
the crop or becomes weakly or strongly adsorbed onto Al, Fe and Ca surfaces. With the 
application of P, available soil P content increases as a function of certain physical and 
chemical soil properties, such as clay, organic C, Fe, Al and calcium carbonate content. 
The continual application of P can result in an increase in soil test P above levels required 
for crop uptake, which has an environmental ramifications. 
Phosphorus Accumulation 

In many parts of the world, concern and research focuses on manure application, 
where the amount of P added often exceeds crop removal rate on an annual basis. Many 
areas with intensive confined animal operations, such as the Netherlands, Belgium, north- 
eastern USA and Florida, now have soil P levels that are of environmental rather than 
agronomic concern (Sharpley et al., 1994b). In 1994, Kingery et al. (1994) reported P 
leaching to a depth of -60 cm in tall fescue pastures in the Sand Mountain region of 
northern Alabama that had received long term-application (15-28 yr) of poultry litter. 
Soil test P (Mehlich I) values in topsoils were extremely high (-230 mg/kg) relative to 
optimum values for crop production in this region (25 mg/kg) (Cope et al., 1981). 
Similarly, Eghball et al. (1996) measured soil test P (Olsen P) in the profile of a Tripp 
very fine sandy loam (a coarse-silty, mixed, mesic Aridic HaplustoU) that had received 
long-term (>50 yr) application of cattle feedlot manure and/or fertilizer P. Crops grown 
included sugarbeet, potato, and com. Increases in soil test P were reported and the 
increases were associated with P leaching to -75 cm with fertilizer P (superphosphate) 
and to -1.0 m for manure or manure plus fertilizer P. Mozaffari and Sims (1994) 
measured soil test P (Mehlich I) values with depth in cultivated and wooded soils on 
farms on a coastal plains watershed dominated by intensive poultry production and 



frequent applications of poultry litter, and observed P leaching to depth of -60 to 75 cm 
in agricultural fields and a high soil test P values in topsoils relative to those considered 
optimum for most agronomic crops (25 mg/Kg) (Sims and Gartley, 1996). In North 
Carolina, King et al. (1990) examined the effect of 1 1 years of swine lagoon effluent 
application on P distribution within the profile of a Paleudult soil used for coastal 
bermuda grass pasture, and reported soil test P (Mehlich I) values much greater than 
required for crop production (225-450 mg/kg as a function of effluent rate, vs. an 
optimum soil test value of -20-25 mg/kg). Soil test P at the 15 to 30, 30 to 45, 45 to 60, 
and 60 to 75 cm depths was <5 mg/kg in nearby unfertilized pasture. However, at the 
same depths, soil test P was about 120, 75, 25, and 5 mg/kg at the lowest effluent rate 
(335 kg N/ha per year) and 350, 175, 125, and 50 mg/kg at the highest effluent rate (1340 
kg N/ha per year). 

The same trend of P accumulation and leaching has also been shown in Florida 
which has intensive agricultural activity, humid climate, frequent heavy rainfall, and 
widespread use of irrigation and drainage. Several studies have shown the extent of P 
leaching that can occur in deep, sandy soils. For example, a study by Wang et al. (1995) 
found that high levels of P could be leached from surface (Ap) horizons of four sandy 
Florida soils heavily loaded with dairy manure despite high pH and abundant Ca^^ in 
solid and solution phases. Total P (TP) ranged from 3 144 to 1 595 mg/kg. Further 
investigation on the composition of the same samples by Harris et al. (1994) showed that 
the dominance of noncrystalline Si and lack of crystalline Ca-P in the intensive area Ap 
horizons constitute an unfavorable environment for P retention in these soils. The 
crystallization of Ca-P may be inhibited by manure-derived component such as Mg, 



10 

organic acids, and Si. Nair et al. (1995) also studied the forms of P in soil profiles from 
dairies of south Florida. The dairies selected were active (still operating at the time of 
sampling) and abandoned (dairies that had not been operating for 4, 12, 18 yr prior to 
sampling). Three components of each active dairy were sampled: intensive areas (areas 
next to the bam where cattle are held immediately prior to milking), pasture (areas used 
for grazing), and forage areas (used for forage production). Their result showed a TP for 
the A horizon ranging from 3028 mg/kg for the active-intensive areas to 2933 mg/kg for 
the abandoned-intensive areas. Total P content of the unimpacted soils (native) was in the 
range of 15-59 mg/ kg for all horizons, with low values observed in the E horizon and 
high values in the Bh horizon. Labile P content (defined as P in sorbed phase, which is 
potenfially mobile and bioavailable) in the Bh horizon of native forage and pasture areas 
were less than 2% of the TP, while up to 10% of TP was found as labile P in surface 
horizons. In intensive areas, up to 40% of the P was in the labile pool. Soil P content 
varied both with soil depth and land use. Total P stored in the soil profile increased with 
intensity of land use, with native unimpacted areas containing 44 g P m"^ (average profile 
depth 99 cm), followed by forage (46 g P m"^; soil depth 94 cm) pasture (102 g P m"^; soil 
depth 1 19 cm) and intensive areas (766g P m"^; soil depth 136 cm). Dairy lagoon effluent, 
though its composition and P content is quite different from dairy manure could also 
elevate the level of P in the soil. Dooley (1996) studied P accumulation and retention in a 
wetland impacted by approximately 20 years of dairy lagoon effluent application and 
showed that the wetland appeared to be exporting P to an adjacent stream. His study 
concluded that in order to accomplish acceptable levels of treatment, the assimilative 
capacity of the wetland must be considered. 



11 

Numerous studies on accumulation of P in soils amended with commercial 
fertilizers and /or organic wastes, including some of the above-mentioned studies, have 
been reviewed recently by Sims et al. (1998). He indicated clearly that the most common 
agricultural situation associated with significant downward movement of P has been the 
accumulation of P to "very high" or "excessive" levels in soils from continuous 
application of organic wastes (manure, litter, and municipal or industrial wastes and 
waste waters). 
Phosphorus Forms and Fractionation 

In order to understand the potential for P transport, P forms have to be examined 
and evaluated to develop an understanding of the stability of P in the soil of the area 
adjacent to the water body. The objectives of P fractionation in general are to provide 
insight into the fate and transformation of P added to soils as fertilizers or manure, 
estimate the availability of P to plants for agronomic purposes, estimate the potential for 
P movement from erosion and through leaching, and provide information regarding the 
interaction between P in sediments and the overlaying water in the case of aquatic 
systems (Graetz and Nair, 1999). Fractionation schemes using various chemical extracts 
have been developed through the years to quantify the different forms of P in soils. The 
underlying assumption here is that inorganic P in soil consists of varying proportions of 
three discrete classes of compounds, namely, phosphates of Fe, Al and Ca, some of which 
could be occluded or enclosed within coatings of Fe oxides and hydrated oxides. These 
chemical P forms are operationally defined and subject to broad interpretations. 
Nevertheless, they offer a convenient means for obtaining significant information on P 
chemistry of soils. For example, through a modification of Hieltjes and Lijklema (1980) 
fractionation method Nair et al. (1995) examined soil phosphorus in soil from dairies of 



12 

south Florida and fractionated it into labile P, inorganic Fe/Al-P, Ca/Mg-P and residual- 
P. This fractionation scheme offered significant information on the forms of P in soil 
profiles from dairies of south Florida. The Hieltjies &, Lijkiema (1980) scheme uses 1 M 
NH4CI to extract loosely bound and labile P. This fraction is believed to contain the water 
soluble portion and the plant- available portion of TP in the sample. Sodium hydroxide is 
the next step in the fractionation procedure. This extract contains both organic and 
inorganic P forms. The inorganic P portion of the extract is believed to be associated with 
Fe and Al, while the organic portion is believed to be fulvic-and humic bound. 
Hydrochloric acid is the third step to remove calcium-bound phosphate. The remaining 
soil can then be digested to measure any residual P. This portion of P is considered to be 
highly resistant, organically bound. 

The forms of P in soil profiles from dairies of south Florida illustrate the fate and 
transport of P in these systems Nair et al, (1995), They identified the P forms in the soil 
profile of differentially manure-impacted soils in the Okeechobee watershed of south 
Florida. All soils were Spodosols, and soils were collected by horizon. A, E, Bh, and Bw. 
Their resuhs showed no statistical differences in the percentage of labile P (NH4CI- 
extractable P), the P that would most likely move from A horizon of the various 
components. More P will be lost from the heavily manure-impacted intensive areas with 
high TP values, than from the less impacted pasture, forage and native areas. They also 
observed that the P would continue to be lost from dairies that have been abandoned for 
considerable period of time. The P that leaves the surface horizon might be lost through 
surface and subsurface drainage, and the portion that reaches the spodic (Bh) horizon will 
be held as Al- and Fe-associated P, either in the inorganic or in the organic fraction. The 



13 

high percentage of HCl-extractable P (Ca- and Mg-associated P) in the A horizon of the 
intensive dairy component was also of potential concern. This P could be continuously 
extracted by NH4CI or by water (Graetz and Nair, 1995), suggesting that about 80% of 
the total soil P had the potential to move eventually with drainage water into Lake 
Okeechobee. 

Fractionation of P forms has been particularly useful in understanding the 
transformation of P added to soil, either in inorganic or organic amendments such as 
manures. Zhang and Mackenzie (1997) used P fractionation and path analysis to compare 
the behavior of fertilizer and manure-P in soils. Their results showed that P behaves 
differently when added as manure, compared to inorganic fertilizer, which may affect the 
depth of P movement through the soil profile. Simard et al. (1995) reported that a 
significant portion of the P moving downward in soils receiving substantial amounts of 
animal manure accumulated in labile forms such as water-soluble, Mehlich-3, and 
NaHCOs extractable P forms. Eghball et al. (1996) found that P from manure moved 
deeper in the soil than P fi-om chemical fertilizer in long term (>50 yr) studies. 
Phosphorus Retention 

P retention in soils is a result of many soil physical and chemical properties, such 
as mineralogy, clay content, pH and organic matter content, that influence the P solubility 
and adsorption reactions. Consequently, these soil properties also affect solution P 
concentration., P availability and recovery of P fertilizer by crop (Tisdale et al., 1993). 
The term fi-equently used to describe surface adsorption and precipition reactions 
collectively is P fixation or retention. The term adsorption and chemisorption also have 
been used to describe P reaction with mineral surfaces, where chemisorption generally 
represents a greater degree of bonding to the mineral surface. The term sorption has been 



14 

used to describe adsorption and chemisorption collectively. Adsorption is the preferred 
term (Tisdale et al., 1993). There is considerable evidence suggesting that, P retention is a 
continuous sequence of precipitation and adsorption. With low-solution P concentration, 
adsorption probably dominates, while precipitation reaction proceed when the 
concentration of P and associated cations in the soil solution exceeds that of the solubility 
product (Ksp) of the mineral. Mineral solubility represents the concentration of ions 
contained in the mineral that is maintained in solution. Each P mineral will support a 
specific ion concentration which depends on the solubility product of the mineral. The 
most common P minerals found in acid soil are Al-and Fe-P minerals, while Ca-P 
minerals predominate in neutral and calcareous soils. But, the specific P minerals present 
in the soil and the concentration of solution P supported by these minerals are highly 
dependent on solution pH (Tisdale et al., 1993). 

Phosphorus sorption may be determined by single point adsorption isotherm or 
multi-point adsorption isotherms and can be described by several different adsorption 
equations; all are based on the fundamental equation: 

q = f(C) 

where q is the quantity of P adsorbed at P concentration C. 

One of the earliest equations used in soil studies is the Freundlich equation, 

q= ac*" 

The amount of P adsorbed per unit weight of soil is q, c is the P concentration in 
solution, and a and b are constant which vary from soil to soil. The Freundlich equation 
was introduced as a purely empirical equation. It implies that the energy of adsorption 
decrease exponentially with increasing saturation of the surface. However, no maximum 



15 

capacity of adsorption can be calculated because the amount of adsorption increases with 
the adsorbing ions in the solution (Yuan and Lucas, 1982). Therefore, it applies well only 
over a limited concentration range of ions to be adsorbed. For this reason, the Langmuir 
equation, which is based on the assumption that adsorption is on localized sites, the 
energy of adsorption is constant, and the maximum adsorption possible corresponds to a 
complete monomolecular layer is often preferred for the description of soil P adsorption. 
In its linear form, the regression line would provide a means to calculate not only the 
maximum adsorption but also a constant which is assumed to be related to the bonding 
energy of the surface for P in solution. The equation describes a finite limit to adsorption 
so that a maximum value may be obtained (Yuan and Lucas, 1982). Although there are 
several linear forms, they are all derived from the basic expression; 

q = kbc/(l+kc) 
where q and c are as in the Freundlich equation, b is the "P adsorption maximum", and k 
is a constant related to bonding energy. 

The Temkin equation, as proposed for use in soil-P system by Bache and 
Williams (1971), also implies that the energy of adsorption decreases as the amount of P 
sorbed increases. In the middle range of P sorption, the equation may be expressed as 

q^ = (RT/B)ln Ac 
where A and B are constant and b, c and q are as in the Langmuir equation. All three 
equations require that equilibrium conditions exist, a state that is rarely achieved in soil-P 
adsorption studies. Another assumption common to the three equations is that the 
adsorption is reversible; however some portion of the P adsorbed by soil is irreversibly 
adsorbed. Despite these and other disadvantages, the three equations have been useful in 



16 

describing the relationship between c and q over limited range of concentrations (Olsen 
and Khasawneh, 1980). 

In Florida, the P retention characteristics of upland and wetland soils and stream 
sediment in the Lake Okeechobee Watershed (maximum P retention capacity [Smax] and 
equilibrium P concentration [EPCo]) have been a point of interest for several studies 
(Reddy et al., 1996). The Smax of Bh horizons was about three to four times higher than 
the surface A and E horizons. High EPCo (equilibrium concentration when net adsorption 
equal zero) values for soils in the A and E horizons suggest poor retention capacity, while 
low EPCo values of the Bh horizon indicate strong affinity for P. The Smax was found to 
be highly correlated with oxalate-extractable Fe and Al, and total carbonate of the soil. 
Oxalate-extractable Fe and Al represent amorphous and poorly crystalline forms. Many 
soils effectively retain P due to the presence of mineral components with high surface 
affinity for orthophosphate. However, movement of P from dairy farms to aquatic 
systems does occur under certain conditions, and has been linked to eutrophication of 
surface water. This movement may be related to erosion or to subsurface transport. 
Subsurface transport of P can be significant in sandy soils due a paucity of P-retaining 
components (Reddy etal, 1996). 

Sandy soils generally retain less P than finer textured soils because of a deficiency 
of mineral components having surface affinity for orthophosphate. In a 1982 study by 
Yuan and Lucas pertaining to the retention of phosphate by thirty Florida sandy soil as 
evaluated by adsorpfion isotherms showed that the simple linear Freundlich equation 
describe the P adsorption properties of sandy soil more successfully than the Langmuir 
equation. The adsorption maximum values obtained from the Freundlich equation were 



17 

correlated with soil properties. A significant relationship was found with clay content but 
not with double acid extractable Al, Fe, Ca and Mg, individually or combined. However, 
for soils with a pH below 5.5, the adsorption maximum had a significant relationship with 
extractable Al. A study on P retention as related to morphology and taxonomy of sandy 
coastal plain soil materials by Harris et al (1996) distinguished between two groups of 
uncoated Quartzipesamments (< 5% silt-plus-clay); those having "clean" (coating-free) 
and "slightly-coated" grains, All clean samples readily desorbed P regardless of origin or 
amount adsorbed. Sand-grain coatings significantly enhanced P adsorption and resistance 
to desorption. Thus, clean sands pose a greater hazard for P leaching than sands with 
grain coatings. Clay content was closely related to P adsorption, but silt content was not. 
The P-retention distinction between clean and other Quartzipsamments is more marked 
than uncoated vs. coated family criterion. The distinction between clean and other sandy 
materials was more discrete and consistent for P desorption behavior than for adsorption. 
A P-adsorption measurement such as the RPA (Rapid Chemical Assessment of Relative 
Phosphorus Adsorption [single-point isotherm]) would provide a reasonably valid 
assessment of P retention for slightly-coated and coated sand materials if appropriately 
calibrated. The RPA effectively arrayed sandy Florida soil samples with respect to 
relative P adsorption. A single-point isotherm could effectively index these sandy 
materials. It does not directly provide values for maximum P adsorption, but it closely 
relates to such values derived from P adsorption isotherm for the same sandy soil studied 
(Harris etal, 1996). 

A recent study by Nair et al. (1998) conducted on Spodosols in the Lake 
Okeechobee basin to evaluate the P retention capacity of manure impacted Bh horizons 



18 

under aerobic and anaerobic conditions found that a high watertable decreased the P 
retention for the majority of the soils in that study. High manure-impacted areas have Bh 
horizons with high P concentrations as a result of P movement from the surface A 
horizon through the eluted E horizon. The P appeared to be temporarily retained and 
could be released upon prolonged contact with water. Another study by Nair et al. (1999) 
of Spodosols in the same basin showed that the surface A and E horizons of manure- 
impacted soils had essentially no sorbing capacity while the Bh (spodic) and Bw horizons 
had mean S^ax values 430 and 385 mg/kg, respectively. The P retention characteristics of 
these soils were determined by using both single-point (1000 mg P/kg or 100 mg P/L) 
and traditional Langmuir isotherms. Phosphorus sorption values using a single high P 
solution had approximately a 1 : 1 relationship with values obtained for the maximum 
retention capacity (S^ax) obtained from Langmuir isotherms. 

In response to the fact that the P sorption capacity of soils is not unlimited, and 
based on documented accumulation and leaching of P in soils of areas dominated by 
concentrated animal production, a new approach to sorption capacity was developed. The 
concept of Degree of P Saturation (DPS) is based on the fact that the potential for soil-P 
desorption increases as sorbed P accumulates in soil (Van der Zee et al., 1987; 
Breeuwsma and Silva, 1992). Degree of P saturation is defined as the ratio of extractable 
P to the sum of extractable Fe and Al expressed as an percentage. The critical DPS 
threshold has been defined as the saturation percentage that should not be exceeded to 
prevent adverse effect on ground water quality with the specific goal that the phosphate 
concentration in the ground water should not exceed 0.01 mg/L of orthophosphate at the 
level of the mean high watertable (Breeuswma et al., 1995). A critical DPS value of 25% 



19 

has been used in the Netherlands to determine the surplus of P that can be applied to 
varying soil types before P saturation, and thus significant P export in subsurface runoff, 
can be expected to occur. 

Operationally, DPS is defined as oxalate-extractable P divided by the phosphate 
sorption capacity of the soil that is estimated from equations including oxalate- 
extractable Fe and Al (Breeuwsma et al., 1995). 



DPS = Extractable soil P X 100 

P sorption maximum 



The extractable P (Pox ) is determined by extraction by 0.2 M ammonium oxalate 
buffered to pH 3.0. Phosphorus sorption capacity is determined by standard P adsorption 
isotherms or estimated by oxalate-extractable Al (Alox ) and Fe (Fcox ) and the DPS 
expressed as: 

Pox 

DPS = X 100 

a (Fcox + AJox) 

The saturation factor a, as defined by Sjoerd et al. (1988), is the ratio of the 
amount of P that is sorbed in laboratory experiments and the P already present as Pox to 
(Feox+ Alox). Thus a is a variable that allows comparison of different soils with respect to 
P saturation, and the result will then be normalized with respect to the reactive soil 
constituents. However, as pointed out by Sjoerd et al. (1988), the proportionality factor a 
is both concentration and time dependent. Pautler and Sims (1998) used an a value 
ranging from 0.4-0.6 for soils of the Atlantic coastal plain. 



20 

An added advantage of the DPS approach is that it not only describes the potential 
for P release from soil but also indicates how close the P-sorption sites of a soil are to 
being saturated (Sibbesen and Sharpley, 1997). Citing data from numerous studies in the 
Netherlands, showed that more than 80% of the soils in a watershed with intensive 
livestock production were saturated with P (Breeuwsma et al., 1995). During winter 
months, when groundwater discharge to surface waters was highest, concentration of TP 
in the shallow groundwater to exceeded surface water quality standards (0. 15 mg/L of 
TP). Lookman et al. (1995) applied the DPS approach to a 700 Km^ area (primarily 
grassland used for intensified animal agriculture) in northern Belgium. Based on a critical 
DPS value of 25%, they estimated that >75% of the soils were considered to be saturated 
with P to the depth of the highest average groundwater table. Lookman et al. (1996) 
showed that the DPS of the same soils of their 1995 study (Lookman et al., 1995) at the 
0-30 cm depth was highly correlated with soluble P in these soils. Sharpley (1995) also 
found a single relationship (r^ of 0.86) describing the concentration of dissolved 
phosphate (DP) as a function of P-sorption saturation for ten soils ranging from sandy 
loam to clay in texture. He used Mehlich-3 as extractable soil P and the Langmuir P- 
sorption maximum as P-sorption capacity in his calculation of P-sorption saturation. 
Comparison of short-and long-term sorption kinetics in Atlantic coastal plain soils 
showed that the potential for P loss from over-fertilized soils can be improved by a 
knowledge of the degree of P saturation of the soils (Pautler and Sims, 1 998). 
Downward P movement 

Loss of P from land can occur in three ways; as water-soluble and/or particulate P 
in surface runoff, as water-soluble and/or particulate P in subsurface runoff (leaching), 
and as water-soluble and/or particulate P in flow to groundwater, referring to P picked up 



21 

by water that passes to the water-table and which is subsequently discharged to streams, 
rivers or lakes as seepage (Ryden et al., 1973). Phosphorus leaching has normally been 
considered to be inconsequential in most soils, but recent studies show that there are 
combinations of agriculture management practices, soil proporities, and climatic 
conditions that can result in significant accumulation in subsoils. Whether or not P that 
leaches into subsurface horizons is later transported to water bodies depends on the depth 
of leaching and the hydrological connections of the watershed (Sims et al., 1998). As 
mentioned above in the section on P accumulation, numerous studies on accumulation of 
P in soils amended with commercial fertilizers and/or organic wastes have been reviewed 
recently by Sims et al., (1998). This indicated clearly that the most common agricultural 
situation associated with significant downward movement of P has been the accumulation 
of P to "very high" or "excessive" levels in soils from continuous application of organic 
wastes (manure, litter, and municipal or industrial wastes and waste waters). Studies by 
Kingery et al. (1994), Eghball et al. (1996), MozafFari and Sims (1994), and King et al. 
(1990) report P leaching to~75 cm depending upon other factors such as soil type and P 
accumulated in the surface horizon. Furthermore, Eghball et al. (1996) suggested a 
greater downward mobility for organic forms of P. Previous studies from Florida also 
illustrated the extent of P leaching that can occur in deep, sandy soils. One of the eariiest 
studies in Florida was of Bryan (1933) who reported P leaching to depths of at least 90 
cm in heavily fertilized citrus groves of varying ages. Humpherys and Pritchett (1971), in 
their study of six soil series in northern Florida, 6 to 10 years after applying 
superphosphate, reported extensive P leaching and subsequent accumulation in the spodic 
horizon of a Leon fine sand and that all fertilizer P had leached below a depth of 50 cm in 



22 

Pomello and Myakka soil series. A study by Wang et al. (1994) found that high levels of 
P could be leached from surface (Ap) horizons of four sandy Florida soils heavily loaded 
v^ith dairy manure despite high pH and abundant Ca^"" in solid and solution phases. 
Graetz and Nair (1995), Nair et al. (1995), Nair et al. (1998), and Nair et al. (1999), in a 
series of studies on Spodosols in the Lake Okeechobee basin of Florida, concluded that P 
that leaves the surface (A) horizon might be lost through surface and subsurface drainage, 
and the portion that reaches the spodic (Bh) horizon will be held as Al- and Fe-associated 
P, either in the inorganic or in the organic fraction. The high percentage of HCl- 
extractable P (Ca- and Mg-associated P) in the A horizon of the intensive dairy 
component was also of potential concern. This P could be continuously extracted by Nil, 
CI or by water, suggesting that about 80% of the total soil P had the potential to move 
eventually with drainage water into Lake Okeechobee (Graetz and Nair, 1995). 

Recently, Sims et al. (1998) reviewed some current research on P leaching and 
loss in subsurface runoff in Delaware, Indiana, and Quebec and concluded that the 
situation most commonly associated with extensive P leaching, and thus the increased 
potential for P loss via subsurface runoff, has been the long-term use of animal manures. 
Manure Management 

Developing manure management plans that are agronomically, economically, and 
environmentally sound is a challenge because issues like accelerated eutrophication, P or 
N limitation, transport mechanisms, source management, soil P level, environmental soil 
testing for P, manure management and land application of manure have to be considered. 
This review of the literature shows the urgent need for research especially in areas of 
intensified dairy production and deep, coated sandy soil. Many factors can be involved in 
developing an environmentally sound plan for manure management. Animal manure can 



23 

be a valuable resource if it can be integrated in cost effective best management practices. 
Uptake of nutrients by agronomic crop sequenced over time is an effective, economical, 
and environmentally sound means of nutrient recovery, especially if the cropping system 
met the environmental concerns. The environmental concerns can be meet by maximizing 
nutrient uptake by the crops while meeting the need of dairy producers. 

A recent two years study on the use of dairy manure effluent in a rhizoma 
(perennial) peanut based cropping system (French et. al. 1995) suggests that, if N 
pollution is the major concern in a particular area, then the PP-R cropping system (year- 
round perennial peanut and rye) would be a good choice since it performed as well or 
better than the C-FS-R (corn, forage sorghum, and winter rye) and C-PP-R (com planted 
into a perennial peanut sod, perennial peanut, and rye) systems. However, if P is the 
major concern, the C-FS-R and C-PP-R systems would be better choices. The C-FS-R 
and C-PP-R systems were superior to the PP-R rotation in P removal values. Though P 
level in perennial peanut forage were generally higher than those in com and forage 
sorghum, they were not high enough to compensate for the much lower annual dry matter 
yield of the perennial peanut system. 

Dissertation Format 
The subsequent chapters in this dissertation were prepared as individual 
manuscripts. In this chapter, a general introduction, statement of the problem, review of 
literature, and research objectives were presented. In chapter 2, the accumulation of P in a 
sandy soil receiving dairy waste effluent was investigated. In chapter 3, the forms and 
fractionation of P in the area under study were examined. In chapter 4, the retention 
capacity of the soil was evaluated. In chapter 5, downward P movement was examined. 



24 

In chapter 6, plant uptake of the cropping systems under study was investigated. Chapter 
7 provides a summary and conclusion of results presented in previous chapters. 



U::vV: .jr,^^ ^O^^l f 



\ i 



-■"\ • ' ■ i ? 



CHAPTER 2 
PHOSPHORUS ACCUMULATION IN A SANDY SOIL RECEIVING DAIRY 

WASTE EFFLUENT 



Introduction 
The number of soils with plant-available P exceeding the levels required for 
optimum crop yield has increased in areas of intensive agriculture and livestock 
production (Sims, 1992; Snyder et al., 1993). In many parts of the world, concern and 
research focuses on manure application, where amounts of P added often exceeded crop 
removal rate on an annual basis. Many areas with intensive confined animal operations, 
such as the Netherlands, Belgium, north-eastern USA and Florida, now have soil P levels 
that are of environmental rather than agronomic concern (Sharpley et al., 1994b). In 
1994, Kingery et al. (1994) reported P leaching to a depth of -60 cm in tall fescue 
pastures in the Sand Mountain region of northern Alabama that had received long term- 
application (15-28 yr) of poultry litter. Soil test P (STP) (Mehlich I) values in topsoils 
were extremely high (-230 mg/kg) relative to optimum values for crop production in this 
region (25 mg/kg) (Cope et al., 1981). Similarly, Eghball et al. (1996) measured STP 
(Olsen P) in the profile of a Tripp very fine sandy loam (a coarse-silty, mixed, mesic 
Aridic Haplustoll) that had received long-term (>50 yr) application of cattle feedlot 
manure and/or fertilizer P. Crops grown included sugarbeet, potato, and com. Increases in 
STP were reported and the increases were associated with P leaching to -75 cm with 
fertilizer P (superphosphate) and to -1.0 m for manure or manure plus fertilizer P. 
Mozaflfari and Sims (1994) measured STP (Mehlich I) values with depth in cuhivated and 



25 



26 

wooded soils on farms in a coastal plains watershed dominated by intensive poultry 
production and frequent applications of poultry litter, and observed P leaching to depth of 
-60 to 75 cm in agricultural fields. STP ( Mehlich I)values in these soils were very high 
in topsoils relative to those considered optimum for most agronomic crops (25 mg/kg) 
(Sims and Gartley, 1996). In North Carolina, King et al. (1990) examined the effect of 1 1 
years of swine lagoon effluent application on P distribution within the profile of a 
Paleudult used for coastal bermudagrass pasture. They reported STP (Mehlich I) values 
much greater than required for crop production 225-450 mg/kg vs. an optimum soil test 
value of -20-25 mg/kg. Soil test P at the 15 to 30, 30 to 45, 45 to 60, and 60 to 75 cm 
depths was <5 mg/kg in nearby unfertilized pasture. However at the same depths, STP 
was about 120, 75, 25, and 5 mg/kg at the lowest effluent rate (335 kg N/ha per year) and 
350, 175, 125, and 50 mg/kg at the highest effluent rate (1340 kg N/ha per year). 
Phosphorus loading from dairy lagoon effluent to soils in the Lake Okeechobee Basin, 
Florida resulted in significant accumulation of P. In some cases; P concentrations were 
about 50 times that of unimpacted areas (Graetz and Nair, 1995). 

A considerable body of research now shows that STP levels influence the amount 
of P in runoff water and subsurface drainage (Pote et al., 1996 ; Sharpley et al., 1977; 
Heckrath et al., 1995). Therefore, STP could help identify areas of potential losses of P. 

This study was initiated to investigate the accumulation of P in the soil profile 
during application of dairy waste effluent to two cropping sequences at two N rates in a 
deep sandy soil. 



27 

Materials and Methods 
Experiment Location and Design 

The study was located at the North Florida Holstein Dairy facility, which is two 
miles south of Bell, Florida. A randomized block design containing three blocks and 
arranged as a split plot was used as the experimental design. Main plots were N loading 
rates and subplots were cropping systems. Subplot area was 232 m^. Dairy waste effluent 
was used as the N source. The N application rates were 448 and 896 kg/ha/yr which 
correspond to P loadings of 112 and 224 kg/ha. The cropping systems were com- forage 
sorghum-rye and perennial peanut-rye. 
Soil Selection and Sampling 

The soil was mapped as a Kershaw sand (sandy, thermic, uncoated Typic 
Quartzipsamments). Soil profile samples (0-15, 15-30, 30-45, 45-60, 60-80, 80-100 cm) 
were collected fi-om each treatment in 1996 (prior to effluent application) and in 1997 and 
1998 (after effluent application). Soil fi-om three profiles in each subplot was collected, 
composited, mixed thoroughly and a 1-kg subsample was brought to the laboratory for 
analysis. Soil samples were air-dried and sieved (2mm) prior to analysis. Soil samples 
were also collected in a similar manner from an adjacent native area believed to be 
unimpacted by manure or fertilization application. 
Soil Characterization 

Texture was determined using the pipette method (Day, 1965). Total phosphorus 
(TP) was determined by ashing 1.0 g of soil for 3 hours and then solubilizing with 6 M 
HCl (Anderson, 1976). Double-acid (Mehlich I)-extractable P, Al, Fe, Ca and Mg were 
obtained with a 1:4 soil/double acid ratio (Mehlich, 1953). Phosphorus (P) in solution 
was analyzed by the molybdenum-blue method (Murphy and Riley, 1962). Soil pH was 



28 

determined on 1:2 soil/water ratio, and the organic carbon content of the air-dried 
samples was determined by combustion (Broadbent, 1965). 
Effluent Application and Characterization 

Effluent was taken directly from the dairy waste pond on the farm in which the 
manure flushed from the milking parlor and feed barn is collected. The effluent was 
applied to the experimental area through a center pivot irrigation system. The annual 
application of effluent ranged between 355 to 500 mm depending on N application rate 
and the concentration of N in the effluent. The average annual concentration of TP 
ranged from 56 mg/L in 1996 to 49 mg/L in 1998, and the soluble reactive phosphorus 
(SRP) from 44 in 1996 to 47 mg/L in 1998 (Table 2-1). v : 



Table 2-1. Average annual concentrations (mg/L) of ammonium nitrogen (NH4 -N), total 
Kjeldahl nitrogen (TKN), soluble reactive P (SRP), and total P (TP) in effluent applied to 
the study site. Numbers in parentheses are standard deviations. 





NH4-N 


TKN 


SRP 


TP 


YEAR 
























mg/L 




1996 


172(48) 


258(84) 


44(14) 


56(20) 


1997 


176(40) 


302(75) 


44(12) 


55(18) 


1998 


192(51) 


280(69) 


47(6) 


49(20) 


Statistical Analv 


sis 









Data analyses were done using SAS program (SAS Institute Inc. 1985) (PROC 
MIXED) procedure (SAS Institute Inc. 1992). The PROC MIXED procedure was 
selected based on the fact that it is designed for a mixed effect model where random 



' ,. -i } .■. t. 



■J- »■, <f -5 ■?-> •■ ■•■ 



29 -. :-: *v,-- 

terms are incorporated into inference from the outset. Contrast, least square means and 
estimates of linear combinations are reported with correct standard errors. The GLM 
(General linear Model) which is designed for a fixed effect model, with allowance for 
certain adjustments in the presence of random terms, needs special attention to be given 
to least square means and contrast since their standard errors are not necessarily correct. 
This is true, for example, for split-plot design as is the case for the experimental design in 
this study (Schabenberger, 1996). 

The main difference between of PROC MIKED and PROC GLM is that PROC 
MIXED estimation of variance is based on maximum likelihood while PROC GLM is 
based on method of moments estimation (ANOVA method) of solving expected mean 
squares for the variance components (Schabenberger, 1996). Another advantage of PROC 
MIXED is that it allows data that are missing at random while PROC GLM requires 
balanced data, and ignore subjects with missing data (Wolfmger and Chang, 1996). This 
criteria for PROC MIXED was of interest in handling the analysis of this study. This 
study is a component of a larger project, which include three main treatments (effluent 
application rate) and five cropping systems as a sub treatments in a split plot design with 
the aim of comparing their effectiveness in nutrient recovery and maintenance of 
acceptable levels of N and P in ground water. However, for the purpose of this study, two 
effluent application rate and two cropping systems were selected under the original 
experimental design. The selection of PROC MIXED to analyze data of the study offered 
a means of dealing with unbalanced data. The model used in the analysis included: date, 
block, rate, crop, and depth and their interactions such as date*rate, date*crop, crop*rate, 



30 

date*depth, rate*depth, crop*depth, date*crop*rate, date*rate*depth, date*crop*depth, 
crop*rate*depth, and date*crop*rate*depth. 

Results and Discussion 

The study site soil was mapped as Kershaw sand (sandy, thermic, uncoated Typic 
Quartzipsamments) in the Gilchrist County soil survey report (Soil Survey Staff, Gilchrist 
County, Florida, 1973). Since the publication of the report, the criterion for coated vs. 
uncoated family placement has been changed for the USDA soil taxonomic system (Soil 
Survey Staff, 1999). The sandy materials sampled in this study would meet the criterion 
for coated (5 percent sih plus 2 times the clay content), based on the particle size analysis 
(Table 2-2). Also, some auger borings to 2 m revealed spodic horizons which indicated 
inclusions of Spodosols, and dark colors in the surface horizon in some areas qualify it to 
be an Umberic epipedon, which would result in classification as an Inceptisol (Umbrept) 
rather than a Psamment. Nevertheless, the soil was consistently sandy and similar to 
Kershaw sand with respect to use and management. ^ .< *. '.,'/•■ • 
Soil Properties Prior to Effluent Application :' f ' "" » * ' --/ ' . 

The soil from the study site prior to effluent application had a different chemical 
composition than a soil samples from a native site (Table 2-3). Double acid (Mehlich I)- 
extractable elements and TP concentrations for the study site prior to the application of 
effluent were higher than the concentrations in soil from the native site (Table 2-3). For 
example, Mehlich I-extractable Ca for the study site ranged from 968 mg/kg at the 
surface horizon to 75 mg/kg at the lower depth of the profile (100 cm). Comparable 
values for the native site were 12 and 4 mg/kg, respectively (Table 2-3). Differences in 
Ca and Mg content between the native site and the study site prior to the application of 



31 



Table 2-2. Selected characteristics of typical Kershaw sand (Soil Survey Staff, Gilchrist 
County, Florida, 1973) compared to the study site. 



Location 


Horizon 


Depth 


pH 


Org.C 


SAND 


SILT 


CLAY 






cm 
















% 




Native 


A 


0-18 


4.8 


0.99 


96.5 


1.1 


2.4 \ 




CI 


18-76 


5.0 


0.39 


96.1 


1.8 


2.1 




C2 


76-147 


4.9 


0.16 


97.3 


0.4 


2.3 




C3 


147-203 


5.0 


0.10 


96.1 


2.0 


1.9 


Studv Site 


Al 


0-15 


6.2 


1.54 


93.1 


4.8 


2.0 




A2 


15-30 


6.0 


0.79 


94.1 


4.2 


1.7 




CI 


30-45 


6.5 


0.71 


95.3 


3.1 


1.6 




CI 


45-60 


6.5 


1,58 


95.0 


3.8 


1.2 




C2 


60-80 


6.6 


0.54 


95.4 


2.7 


1.9 




C2 


80-100 


6.5 


0.43 


95.9 


2.4 


1.7 



32 



Table 2-3. Mehlich I-extractable elements concentrations and total P (TP) in "native" soil 
(n = 3 profiles) and study site (n = 12 profiles) soil profiles prior to beginning of the 
study. 



Location Depth Ca Mg Al Fe P TP 
cm 

(mg/kg) 

Native 0-15 11.7 1.9 267 18.4 47 214 

15-30 5.1 1.1 317 20.7 52 270 

30-45 6.0 0.8 330 19.1 39 Ml 

45-60 4.7 0.7 337 16.3 36 li4 

60-80 4.7 0.8 308 16.3 39 181 

80-100 4.1 0.7 280 14.3 33 173 



Studv 


0-15 


968 


115 


301 


23.5 


283 


328 


£ils 


















15-30 


522 


69.3 


280 


22.8 


184 


2S4 



30-45 208 34.1 203 19.6 75 154 

45-60 135 25.7 161 17.9 37 254 

60-80 103 22.9 133 15.9 20 218 

80-100 75 19,2 117 14.5 12 192 



33 



effluent were also reflected in a higher pH in all horizons in the study site. The higher 



pH and organic C in all horizons of the soil from the study site prior to effluent 
application could be attributed to a previous manure application. Dairy manure can 
appreciably elevate not only the P, but also other components in soils (Dantzman et al., 
1983; Wang et al., 1995). The elevated level of P (TP and Mehlich I- extractable P) in the 
soil of the study site indicated a previous manure application (Table 2-3). The site 
appeared to havebeen heavily loaded with animal waste prior to the start of this study (47 
mg/kg Mehlich I-extractable P in the native area vs 283 mg/kg Mehlich I-extractable P in 
study site surface horizon soils). Several studies (Sims, 1992; Snyder et al., 1993; 
Sharpley et al, 1994b; Kingery et al., 1994; Graetz & Nair, 1995; Wang et al., 1995) 
have shown that manure application usually results in an increase in TP, STP and other 
components in soil. 
Effect of Application Rate and Cropping Systems 

Statistical evaluation of TP data (Table 2-4) shows that date and depth were 
significant at the 0.0001 probability level, but date*depth and date*crop was also 
significant at 0.0001. However, neither the single effect of crop (cropping system), nor 
the rate (effluent application rate) was significant. Therefore, a higher level of 
significance such as date*depth and date*crop will be reported and interpreted, when it 
was appropriate. Means comparison was done when there was a significant interaction by 
SAS code (pdiff) for differences between least squares means (LSM). 

Total P (TP) increased over time (1996 vs. 1998). The effect of date*depth was 
significant (P <0.01) to the depth of 45 cm which reflects a buildup of total P in the soil 
profile. AJso, the effect of date*crop was significant (P<0.01) which might imply a role 



34 



- ■< '.. 



Table 2-4. Statistical evaluation of TP data for the three-year study period. 



Source NDF DDF Tyge IIIF Pir> F 



Date 


2 


96 


36.31 


0.0001 


Block . . 


2 


2 


■ . , 0.46 


0.6867 


Rate 


1 


2 


0.37 


0.6063 


Date*Rate 


2 


96 


2.65 


• 0.0760 


Crop 


1 


44 


0.15 


0.7001 


Date* Crop 


2 


96 


7.47 


0.0010 


Crop*Rate 


1 


44 


1.18 


0.237 


Date*Crop*Rate 


2 


96 


0.83 


0.4412 


Depth 


5 


44 


46.25 


0.0001 


Date*Depth 


10 


96 


15.82 


0.0001 


Rate*Depth 


5 


44 


1.21 


0.3226 


Date*Rate*Depth 


10 


96 


1.14 


0.3395 


Crop*Depth 


5 


44 


0.84 


0.5263 


Date*Crop*Depth 


10 


96 


1.69 


0.0937 


Crop*Rate*Depth 


5 


44 


0.6 


0.6410 


Date*Crop*Rate*Depth 10 96 


1.08 


0.3853 



35 

• ■ 

for the cropping system on P removal. The application of effluent at both rates (448 and 
896 kg N/ha per year) increased TP content of the Sbtl and the increase was dependent on 
the effluent application rate. Total P in the surface horizon increased from 3 12 to 753 
mg/kg at the end of the study under the high application rate (Fig. 2-1) and fi-om 343 to 
485 mg/kg under the low application rate (Fig. 2-2). A higher TP content in soil impacted 
by dairy waste application is common. Graetz and Nair (1995) reported up to 1885 mg/kg 
of TP in the soil surface horizon of dairy intensive areas. 

The application of effluent also had an effect on Mehlich I-extractable P. The 
effect of date*depth (P<0.05) and rate*crop (P<0.0\) on Mehlich I-extractable P were 
significant. During the two year of effluent application, Mehlich I-extractable P 
decreased in the surface horizon but increased in the lower horizons (Fig. 2-3). The 
decrease of Mehlich I-extractable P in the surface horizon and the increase in the lower 
depths of the profile may be attributed to both crop uptake of P and the leaching effect of 
effluent irrigation. Mozaffari and Sims (1994) measured soil test P (Mehlich I) values 
with depth in cultivated and wooded soils on farms impacted by poultry litter 
applications, and observed leaching to depth of- 60 to 75 cm. Soil test P values were 
very high in topsoils relative to those considered optimum for most crops. Also, King et 
al. (1990) examined the effect of 1 1 years of swine lagoon application on P distribution 
and reported soil test P (Mehlich I) values much greater than required for crop production 
(225-450) mg/kg. Soil test P at the 15-30, 30-45, 45-60, and 60-75 cm depth was 120, 
75,25, and 5 mg/kg at the lowest effluent rate (335 kg N/ha per year) and 350, 175, 125, 
and 50 mg/kg at the highest effluent rate (1340 kg N/ha per year). Soil test P (Mehlich I) 
values for soil samples from the study site at 15-30, 30-45, 45-60, and 60-80 cm depth 



36 



.»-.;--. > ^^' 






k.:.,U^ 



V 




I -#-1996 -»-1997 -A- 1998] 

Figure 2-1 . Average total P (TP) concentrations in the soil profile under the high rate 
application prior to application of effluent (1996) and after effluent application (1997 and 
1998). Values are LSM± Std. Error. 



37 



\ -'r' 



100 



200 



IP, mg/kg 

300 400 



500 600 



1996 



1997 



1998 



700 




Figure 2-2. Average total P (TP) concentrations in the soil profile under the low rate 
application prior to application of effluent (1996) and after effluent application (1997 and 
1998). Values are LSM+ Std. Error. 



;:*■ '" ' 



120-L 



38 



100 



DA-P, mg/kg 
150 200 



250 



300 



350 




■DAP 96 
-DAP 98 



Figure 2-3. Mehlich I-extractable P concentrations in the soil profile prior to start of the 
study and after two years of effluent application (1998). Values are LSM± Std. Error. 



*'r- 



-, -?- ,v^i'' *Ti— 



39 

were 197, 93, 55, and 32 mg/kg at the highest rate (896 kg N/ha per year) and 157, 62, 41 
and 22 at the lowest rate (448 kg N/ha per year). These values of soil test P for the soil 
samples from the study site after two year of effluent application should be looked at in 
the context of high Mechlich I extractable P existing prior to the start of the study (Table 
2-3). However, as for the rate*crop effect, Mehlich I-extractable P concentration was 
higher for P-R (perennial peanut-rye) than for C-FS-R (corn-forage sorghum-rye) ' 

cropping system under the high rate application (Fig. 2-4). This finding suggests that the 
C-FS-R (corn-forage sorghum-rye) cropping system may be more effective in P removal 
than the P-R cropping system. Only a slight change in Mehlich I-extractable P between 
the two cropping systems was observed under the low application rate (Fig. 2-5). In spite 
of the suggested higher P removal by the C-FS-R cropping system than the P-R cropping 
system, the level of double acid (Mehlich I)-extractable P concentration in the study prior 
to effluent application is considered to be extremely high relative to the optimum for crop 
production when compared to levels reported by other studies (Kingery et al., 1994; 
Mozaffari and Sims, 1994) and the removal by the cropping systems did not alter the high 
level of STP. Such high level of STP can lead to leaching to a deeper depth in the soil 
profile. 

Although double acid may not extract the total amounts of reactive elements for P 
retention, double-acid (Mehlich I)- extractable P was highly correlated with Ca, Mg, Al, 
and Fe extracted by Mehlich I solution, with 93% of variability explained by this 
relationship (Table 2-5). 



>-vrT5jr^-. 



40 



-I ■ < ' 




120 



-P-R 



-C-FS-R 



Figure 2-4. Mehlich I-extractable P concentrations for cropping systems under the high 
rate effluent application in 1998. Values are LSM± Std. Error. 



41 




120 



■P-R 



C-FS-R 



300 



Figure 2-5. Mehlich I-extractable P concentrations for cropping systems under the low 
rate effluent application in 1998. Values are LSM± Std. Error. 



42 



Table 2-5. Regression equation relating Mehlich I-P to the independent variables Mehlich 
I-Ca, Mg, and Fe. (n-432). 



-^q,y:?:t!?iL Model R 



■tfVWVWWWHIWVV 



M I-P = -56.9 + 0.272 M I-Ca*** - 0.284 M I-Mg* + 0.233 M I-Al*** + 6.932*** 

1.55MI-Fe*** 

***, *, Significant atp< 0.001, and/?< 0.05 respectively. N.S. Not significant. 



Summary and Conclusions 
The soil at the study site has been mapped as Kershaw sand and is considered 
uncoated. However, some coatings are evident based on the color of the sand grain and 
the USDA taxonomic criterion of >5% sih plus (2 times the clay content) for coated 
family placement (Soil Survey Staff, 1999). Sands that retain coating components should 
have a higher affinity to retain P than do bare quartz grains (Harris et al., 1996), a 
criterion that is favorable for this study. The soil at the study site appeared to have been 
heavily loaded with animal waste prior to the start of this study. Mehlich I-extractable P 
in the surface horizon of the native area was 47 mg/kg vs. 283 mg/kg Mehlich I- 
extractable P in the study site surface horizon soils. Mehlich I-extractable P levels in 
topsoils at the study site was high relative to those considered optimum for agronomic 
crops and raise the question about the suitability of the effluent application rates used. 
The effluent application rates selected were based mainly on estimated N removal for the 
forage crops within the cropping systems and experimental purposes outlined in the main 
project objectives. 

The previous application of dairy manure to the study site prior to the start of the 
study resulted also in a higher Ca^^ and Mg*^ content throughout the soil profile 



43 

compared to the native site, although the amount and date could not be established. The 
application of dairy waste effluent at both rates (448 and 896 kg N/ha per year) over a 2- 
year period increased the TP content in the soil profile to the 45 cm depth. The increase 
in TP was significant and dependent on the effluent application rate. The application of 
effluent also had an effect on the Mehlich I-extractable P. The effect of date*depth and 
rate*crop on Mehlich I-extractable P were significant. During the two year of effluent 
application Mehlich I-extractable P decreased in the surface horizon but increased in the 
lower horizons. The decrease of Mehlich I-extractable P in the surface horizon and the 
increase in the lower depths of the profile may be attributed to both crop uptake of P and 
the leaching effect of effluent irrigation. However, as for the rate*crop effect, Mehlich I- 
extractable P concentrations were higher for P-R (perennial peanut-rye) than for C-FS-R 
(corn-forage sorghum-rye) cropping system under the high rate application. This finding 
suggests that the C-FS-R (corn-forage sorghum-rye) cropping system may be more 
effective in P removal than the P-R cropping system. Only a slight change in Mehlich I- 
extractable P between the two cropping systems was observed under the low application 
rate. However, the removal by the cropping systems did not alter the high level of STP 
that already existed. Thus, to prevent an accumulation of excessive P content in the soil 
profile, history of the land, application rate, and cropping systems estimated removal of P 
should be considered. 



CPiAPTER3 
PHOSPHORUS FORMS AND FRACTIONATION IN A SANDY SOIL RECEIVING 

DAIRY WASTE EFFLUENT 



Introduction 
Sequential extraction schemes using various chemical extracts have been 
developed through the years to quantify and fractionate the different forms of P in soils. 
The objectives of P fractionation in general are to provide insight into the fate and 
transformation of P added to soils as fertilizers or manure, estimate the availability of P to 
plants for agronomic purposes, estimate the potential for P movement from erosion and 
through leaching, and provide information regarding the interaction between P in 
sediments and the overlaying water in the case of aquatic systems (Graetz and Nair, 
1999). The underlying assumption here is that inorganic P in soil consists of varying 
proportion of three discrete classes of compounds, namely, Fe, Al and Ca phosphate, 
some of which could be occluded or enclosed within coating of Fe oxides and hydrated 
oxides. These chemical P forms are operationally defined on the basis of reactivity of a 
particular phase in a given extractant and subject to several interpretations. Nevertheless, 
they offer a convenient means for obtaining significant information on P chemistry of 
soils (Nair et al., 1995). Fractionation of P forms has been particularly usefiil in 
understanding the transformation of P added to soil, either in inorganic or organic 
amendments such as manures. Zhang and Mackenzie (1997) used P fractionation and 
path analysis to compare the behavior of fertilizer and manure-P in soils. Their results 
showed that P behaves differently when added as manure, compared to inorganic 



44 



45 

fertilizer, which may affect the depth of P movement through the soil profile. Simard et 
al. (1995) reported that a significant portion of the P moving downward in soils receiving 
substantial amounts of animal manure accumulated in labile forms such as water-soluble, 
Mehlich-3, and NaHCOs extractable P forms. Eghball et al. (1996) found that P from 
manure moved deeper in the soil than P from chemical fertilizer in long term (>50 yr) 
studies. Nair et al. (1995) studied the forms of P in soil profiles fi-om dairies of south 
Florida and illustrated the fate and transport of P in these systems. They identified the P 
forms in the soil profile of differentially manure-impacted soils in the Okeechobee 
watershed, Florida. All soils were Spodosols, and soils were collected by horizon. A, E, 
Bh, and Bw. Their results showed no statistical differences in the percentage of labile P 
(NH4 Cl-extractable P), the P that would most likely move fi-om the A horizon of the 
various components. The labile P form for the A horizon of all dairy components 
averaged 9%. However, more P will be lost from the heavily manure-impacted intensive 
areas with high total P values, than fi-om the less impacted pasture, forage and native 
areas. They also observed that the P would continue to be lost fi-om dairies that have been 
abandoned for a considerable period of time. The P that leaves the surface horizon might 
be lost through surface and subsurface drainage, and the portion that reaches the spodic 
(Bh) horizon will be held as Al- and Fe-associated P, either in the inorganic or in the 
organic fi-action. The high percentage of HCl-extractable P (Ca- and Mg-associated P) in 
the A horizon of the intensive dairy component was also of potential concern. Such P 
could be continuously extracted by NH4CI or by water, suggesting that about 80% of the 
total soil P had the potential to move eventually with drainage water into Lake 
Okeechobee (Graetz and Nair, 1995). 



46. 

Recently, other watersheds in Florida such as the Middle Suwannee River area 
have become the focus of attention. Soils in this area include Entisols; soils lacking 
diagnostic horizons and other features that are specifically defined and required for other 
orders of the USD A taxonomic system (Soil Survey Staff, 1994). Quartzipsamments, the 
only Psamment Great Group which occurs in Florida, are in central northern peninsular 
Florida and most prevalent on well to excessively drained landscapes (Harris and Hurt, 
1999). The sandy nature of Quartzipsamments result in relatively low P retention or 
capacity. Therefore, an understanding of P forms in such soil receiving dairy manure 
effluent application could help identify areas of potential losses of P. 

Materials and Methods 

- a ■ 

Experiment Location and Design ^ . : 

The study site was located at North Florida Holstein Dairy facility, which is two 
miles south of Bell, Florida. A randomized block design containing three blocks and 
arranged as a split plot was used as the experimental design. Main plots were N and P 
loading rates and subplots were cropping systems. Subplot area was 232 m^. Dairy waste 
effluent was used as the N source. The N application rates were 448 and 896 kg/ha/yr, 
which correspond to P loading of 1 12 and 224 kg/ha/yr. The cropping systems were corn- 
forage sorghum-rye and perennial peanut-rye. 
Soil Selection and Sampling 

The soil was mapped as Kershaw (sandy, thermic, uncoated Typic 
Quartzipsamments). Soil profile samples (0-15, 15-30, 30-45, 45-60, 60-80, 80-100 cm) 
were collected in 1996 (prior to effluent application) and in 1997 and 1998 (after effluent 
application). Soil from three profiles in each subplot was collected, composited, mixed 



47 

thoroughly and a 1-kg subsample was brought to the laboratory for analysis. Soil samples 
were air-dried and sieved (2mm) prior to analysis. In addition to soil samples from the 
study site, soil samples were also collected in a similar manner from an adjacent native 
area believed to be unimpacted by manure or fertilization application. 
Fractionation Scheme 

The scheme used to fractionate soil-P was a modification of that of Hieltjes and 
Lijklema (1980) by Nair et al, (1995). A 1-g air dried sample was sequentially extracted 
twice with 25 mL of 1 MNH4CI (adjusted to pH 7.0) with two hours shaking, 0. 1 M 
NaOH with seventeen hours shaking, and 0.5 A/HCl with 24 hours shaking. The 1 : 25 
soil: solution ratio was selected based on preliminary investigations as shown in 
APPENDIX. After each extraction, the content were centriftiged for 15 min at 3620 x g 
and filtered through a 0.45-nm filter. All extractions were carried out at room 
temperature. Residual P was determined by ashing previously extracted soil sample for 
three hours and then solubilizing with 6 MHCl (Anderson, 1976). A 5 mL of the NaOH 
extract was also digested by persulfate-sulfuric acid mixture at 380°C (APHA, 1985) to 
determine moderately labile organic P as the difference between P in digested and 
undigested NaOH extract. NHjCl-extractable P was defined as labile P (Petterson and 
Istvanovics, 1988), NaOH-extractable P as Fe-Al-associated P, and HCl-extractable P as 
Ca-Mg-associated P. Residual P is the P that is not readily removed by any of the above 
chemical extractants. Total phosphorus (TP) was determined by ashing 1.0 g of soil for 3 
hours and then solubilizing with 6MHC1 (Anderson, 1976). Double-acid (Mehlich I)- 
extractable P, Al, Fe, Ca and Mg were obtained with a 1 :4 soil/double acid ratio 
(Mehlich, 1953). Phosphorus (P) in solution was analyzed by the molybdenum-blue 



48 

method (Murphy and Riley, 1962) on a spectrophotometer at wavelength of 880 nm.. Soil 
pH was determined on 1:2 soil/water ratio, and the organic carbon content of the air-dried 
samples was determined by combustion procedure (Broadbent, 1965). Texture was 
determined using the pipette method (Day, 1965). -' ^ ' . "> 

Statistical Analysis 

Data analyses were done using SAS program (SAS Institute Inc. 1985) (PROC 
MIXED) procedure (SAS Institute Inc. 1992). Relationships among parameters were 
evaluated using linear correlation. Multiple regression was used to examine the strength 
of the relationships between parameters. 

Results and Discussion 
Inorganic Fe/Al associated P constituted the major proportion of TP in the soil 
profile of the study site prior to effluent application. As concluded from Chapter 2, the 
soil at the study site appeared to have been heavily loaded with animal waste prior to the 
start of this study, although amount and dates could not be established. Phosphorus 
originally present in the soil profile in 1996 (prior to effluent application) was largely in 
the form of inorganic Fe/Al-associated P, which ranged from 292 mg/kg in the surface 
horizon to 76 mg/kg in lower depth (100 cm) (Table 3-1). These values of Fe/Al- P 
corresponded to 62% and 49% of TP, respectively (Table 3-2). The application of 
effluent increased this fraction to 362 mg/kg in the surface horizon in 1998 with smaller 
increase throughout the soil profile (Fig. 3-1). A comparison of Fe/Al-P mean 
concentration in each depth within the soil profile at the beginning (1996) and end of the 
study period (1998) showed that the increase was statistically significant (F <0.001- 
P<0.05) at the surface and down to the 45 cm depth (Table 3-3). The predominance of 



49 



Table 3-1. P values (mg/kg) in each fraction within a soil depth increment at the 
beginning (1996) and end of the study period (1998) (n = 12 profiles). Values are Least 
Square Means (LSM). 



Depth 


Labile 


Al-Fe 


Ca-Mg 


Residual 


(cm) 






mg/kg 




1 c\c\/- 






vyyyj 






0-15 


85 


1^1 


50 


36 


15-30 


61 


221 


30 


22 


30-45 


36 


146 


21 


16 


45-60 


44 


106 


12 


25 


60-80 


37 


87 


09 


29 


80-100 


37 


76 

1QQ8 


12 


25 


0-15 


97 


1 yyo — 
362 


52 


59 


15-30 


93 


317 


30 


40 


30-45 


80 


205 


17 


28 


45-60 


77 


137 


13 


23 


60-80 


71 


101 


07 


23 


80-100 


69 


81 


07 


19 



50 



Table 3-2. Percentage of P in each fraction within a soil depth increment at the beginning 
(1996) and end of the study period (1998)(n = 12 profiles). Values are Least Square 
Means (LSM). 



Depth 


Labile-P 


Al-Fe-P 


Ca-Mg-P 


Residual-P 


Simiof 


(cm) 






O'^ 




P fraction 
mg/kg 
















1996 






0-15 


18.3 


62.6 


10.8 


8.30 


463 


15-30 


18.4 


65.3 


9.20 


7.10 


334 


30-45 


16.8 


64.5 


10.5 


8.20 


219 


45-60 


24.8 


54.9 


7.00 


13.3 


187 


60-80 


24.3 


52.3 


5.80 


17.0 


162 


80-100 


26.3 


49.4 


8.20 

1 nn<> 


16.1 


150 








lyyo 






0-15 


17.0 


63.5 


9.1 


10.3 


570 


15-30 


19.4 


66.0 


6.2 


8.30 


480 


30-45 


24.2 


62.1 


5.1 


8.50 


330 


45-60 


30.8 


54.8 


5.2 


9.20 


250 


60-80 


35.1 


50.0 


3.5 


11.4 


202 


80-100 


39.2 


46.0 


4.0 


10.8 


176 



51 




■1996 



-1998 



Figure 3-1. Al-Fe-associated P (mg/kg) within the soil profile at the beginning 1996 and 
end of the study period (1998). Values are LSM± Std. Error. 



52 



Table 3-3. Increases in each fraction within a soil depth increment between the beginning 

(1996) and end of the study period (1998). 

Depth Labile Al-Fe Ca-Mg Residual 

(cm) mg/kg mg/kg mg/kg mg/kg 



0-15 


11.75* 


69.76** 


NS 


22.31** 


15-30 


31.62** 


95.69** 


NS 


17.69** 


30-45 


44.55** 


59.03** 


NS 


23.9** 


45-60 


33.38** 


31.0* 


NS 


NS 


60-80 


34.52** 


NS 


NS 


NS 


80-100 


31.44** 


NS 


NS 


NS 



*, ** Significant at the .05 and .001 probability levels, respectively; NS = none 
significant. 



53 

Fe/Al-associated P in surfece horizon and throughout the profile was a reflection of the 
properties of soil and the dairy waste effluent used. The soil at the study site, as 
mentioned in Chapter 2, was classified as coated sand with low clay content, low organic 
matter, pH of 6-6.5, and a higher Mehlich I- extractable P (Table 3-4) in comparison to 
soil from a native area (283 vs. 47 mg/kg). The soil from the native area, which has a low 
content of clay, organic matter, moderately low pH (4-4.5), and a high Mehlich I- 
extractable Al/Fe, compared to the rest of cations in the soil, was a typical example of 
predominance of P retention by Al/Fe oxides. Further more, P fractionation of soil 
samples from the native area showed that up to 62% of TP was in the form of Al/Fe- 
associated P (Tables 3-5 and 3-6). 

The predominance of Al/Fe-associated P in the soil samples from the study site 
was a reflection of its origmal properties, and its increase after effluent application could 
be a consequence of adsorption imder continuous application of a soluble P. The dairy 
waste effluent contained 55 mg P/kg, 78% of which was soluble reactive P (SRP). The 
difference between pH values of soil sample from the study site and native area, and the 
presence of a higher Mehlich I-extractable Ca content in soil samples from the study area 
did not seem to alter the predominance of Al/Fe-associated P in the P fractionation 
scheme. 

Labile-P or easily removable P as defined by (Petterson and Istvanovics, 1988) 
constituted 18-40% of TP in the soil profile of the study site. Prior to effluent application 
in 1996, labile-P ranged from 85 mg/kg in the surface horizon to 37 mg/kg in the lower 
depth (100 cm) (Table 3-1) which corresponds to 26 and 18% of TP, respectively (Table 
3-2). The application of effluent increased this fraction to 97 mg/kg in surface horizon 



54 



Table 3-4. Mean concentration of Mehlich I extractable elements (mg/kg) in the soil 
profile of the study site in 1996 prior to the application of effluent (n = 12 profiles). 

Mehlich I Extractable Elements (mg/kg) 

Depth "Cai Mg Al Fe"""" 



Location (cm) 



0-15 


968 


15-30 


522 


30-45 


208 


45-60 


135 


60-80 


103 


80-100 


75 



Study Site 0-15 968 115 301 23 

69 280 23 

34 203 20 

26 161 1« 

23 133 16 

19 117 14 



* 



> 



55 



Table 3-5. P values (mg/kg) in each of the fractions within a soil depth increment at the 
native site (n = 3 profiles). Values are Least Square Means (LSM). 



Depth 


Labile 


Al-Fe 


Residual 


Ca-Mg 


Sum of P 


(cm) 


P 


P 
mg/kg — 


P 


P 


Fractions 
mg/kg 


0-15 


68 


125 


3 


6 


202 


15-30 


59 


195 


20 


12 


26 


30-45 


58 


166 


4 


13 


241 


45-60 


58 


136 


17 


8 


219 


60-80 


58 


150 


17 


9 


234 


80-100 


61 


149 


16 


9 


235 



Table 3-6. Percentage of P in each of the fractions within a soil depth increment at the 
native site (n = 3 profiles). Values are Least Square Means (LSM). 



Depth 


Labile 


Al-Fe 


Ca-Mg 


Residual 


(cm) 


P 


P 

o/ 


P 


P 




/o 






0-15 


33 


62 


3 


2 


15-30 


20 


68 


4 


7 


30-45 


23 


69 


5 


2 


45-60 


26 


62 


4 


8 


60-0 


24 


64 


4 


7 


80-100 


26 


63 


4 


7 



56 

and 69 mg/kg in the lower depth (100 cm) (Table 3-1) and (Fig. 3-2). Labile-P (Fig. 3-2) 
increased in the surface horizon and throughout the profile over time (1996 vs. 1998) 
with a substantial increase in the lower depth accounting for 40% of TP in 1998 (Table 3- 
2). A comparison of labile-? mean concentration in each depth within the soil profile at 
the beginning (1996) and end of the study period (1998) showed that the increase was 
statistically significant (P <0.001 - P<0.05) at the surface and throughout the profile 
(Table 3-3). The previous mentioned studies by Nair et al. (1995) and Graetz and Nair 
( 1 995) has reported that labile P form for the A horizon of Spodosol in all dairy 
components averaged 9% m a single NH4CI extraction and 1 : 10 soil: solution ratio. The 
higher percentage of labile P form in this study throughout the profile and its substantial 
increase in the lower depth after effluent apphcation is likely due to rapid movement of P 
through the profile. 

In this study, the Ca and Mg-associated P fi-action was the only fraction that 
remained constant and did not show change with the application of effluent over time 
(Table 3-1 and 3-2); (Fig. 3-3) in spite of considerable Mehlich I extractable-Ca content 
throughout the soil profile as shown in Table 3-4. Though the stability of P forms is not 
addressed in this study, Harris et al. (1994) reported an absence of Ca-P minerals despite 
high pH and years of high Ca and P additions in soils from intensive areas of dairies in 
south Florida. The lack of crystalline Ca-P could be related to kinetics, or to a poisoning 
effect of component such as Mg, Si and organic acids in the dairy soil system (Wang et 
al, 1995). The absence of a significant change in the Ca/Mg-associated P pool, in this 
study, could be due to the factors mentioned by Wang et al., 1995 or due to analytical 



57 




120 



■1996 -♦-iggs 



Figure 3-2. Labile P values (mg/kg) within the soil profile at the beginning (1996) and 
end of the study period (1998). Values are LSM± Std. Error. 



58 




Figure 3-3. Ca-Mg associated P values (mg/kg) within the soil profile at the beginning 
(1996) and end of the study period (1998). Values are LSM+ Std. Error. 



59 

limitation. Nair et al. (1995) noticed that the labile P fraction increased if the soil was 
repeatedly extracted with the 1 MNH4CI solution, with a corresponding decrease being 
noted for the HCl P fraction (Ca/Mg-associated P pool). 

Residual-P, the P fraction that is not readily removed by any of the chemical 
extractants, constituted 7 to 17% of TP in the soil profile of the study site in 1996 prior to 
the application of effluent (Table 3-1). This percentage corresponded to 25 and 36 mg 
P/kg, respectively (Table 3-2). The application of effluent increased this fraction to 59 
mg P/kg at the surface horizon in 1998 (Table 3-1) and (Fig. 3-4). However, as a 
percentage of TP this amount constituted 10% of total P (Table 3-2). Bowman et al, 
1998 used both terms resistant P and residual P to mean that pool which is extracted with 
great difficulty, or by difference from the whole when a soil residue yields essentially no 
more acid- and base-extractable inorganic P (Pi) and organic P[ Po, as determined by 
difference (Pt - Pi)]. They reported an average of about 26% of TP as resistant, with the 
more weathered soil containing about 50% resistant P. Nair et al. (1995) studied the 
distribution of P forms of two abandoned dairies (12 and 18 yr) compared with the 
youngest active (8 yr) dairy and reported an increase in Ca/Mg-associated P (61-74 %) 
and a decrease m residual P (20 to 1 1%) in the A horizon of the abandoned dairies in 
south Florida. They related this trend to a possible gradual minerahzation of the residual 
P, if the residual P is primarily recalcitrant organic P. However, the trend of increasing 
residual P content in this study could be related to certain components in the effluent 
used. The fractionation scheme used in this study did not offer a way of fractionating 
residual P into organic and inorganic forms. 

The trend of distribution of different P pools in this two year study was 



60 



10 



P, mg/kg 
20 30 40 




Figure 3-4. Residual-P values (mg/kg) within the soil profile at the beginning (1996) and 
end of the study period (1998). Values are LSM± Std. Error. 



..H 



61 

that Al-Fe- associated P constituted the major proportion of TP in soil profile followed 
by labile-? (prior to effluent appUcation) and that both showed an increase with the 
application of effluent. The increase of labile-? in the lower depth of the soil profile after 
effluent application could be an indication of downward ? movement in the soil profile. 

Previous research in forms of P in soil profile from dau-ies of south Florida by 
Graetz and Nair ( 1 995) reported a predominance of Al/Fe-P (49% of TP) in the A 
horizon of Spodosol soils from nonimpacted areas. However, the predominant form of P 
in the A horizon of highly manure- impacted areas (active dairies for up to 32 years) was 
Ca/Mg-P which reflect the predominance of Ca/Mg in cattle manure in their case. They 
also reported that high percentage of HCl-extractable P (Ca/Mg-associated P) in the A 
horizon of the intensive dairy component was of potential concern. This P could be 
continuously extracted by NH4 CI or by water, suggesting that about 80% of the total soil 
P had the potential to move eventually with drainage water into Lake Okeechobee (Nair 
et al, 1995). 

Summary and Conclusions 
Most of P in the soil profile of the study site (prior to effluent application) 
consisted of Fe/Al-associated P, which accounted for 49-62% of TP. The application of 
effluent resulted in an increase in this fraction throughout the soil profile. Labile-P 
constituted 18-26% of TP in the soil profile of the study site prior to effluent application, 
and the application of effluent increased this fraction significantly up to 40% of TP at the 
lower depth of the profile (100-cm). The increase m labile-P at the lower depth (100-cm) 
of the soil profile after two years of effluent application could be an indication of 
downward P movement in the soil profile. Ca/Mg-associated P was the only fraction that 



/ '\ 



62 

remained constant and did not show change with the appHcation of effluent over time. 
However, the absence of a significant change in Ca/Mg-associated P in this study could 
be due to analytical limitation. Nair et al. (1995) noticed that the labile P fraction 
increased if the soil was repeatedly extracted with the 1 MNH4CI solution, with a 
corresponding decrease being noted for the HCl- P fraction (Ca/Mg-associated P pool). 



CHAPTER 4 
PHOSPHORUS RETENTION IN A SANDY SOIL RECEIVING DAIRY WASTE 

EFFLUENT * f 

■• ■■*. .:':^*'' 

Introduction 

Sandy soils generally retain less P than finer textured soils because of a deficiency 
of mineral components having surface affinity for orthophosphate. Thus, subsurface 
transport of P can be significant in sandy soils due to low surface area or a paucity of P- 
retaining components (Reddy et al., 1996). However, sand-grain coatings could 
significantly enhance P adsorption and resistance to desorption (Harris et al., 1996). 
Phosphorus retention in such soil has been the focus of a number of studies due to its 
relevant environmental consideration in areas of intensified animal-based agriculture 
(Mozaffari and Sims, 1993; Harris et al., 1994; Graetz and Nair, 1995; Nair et al., 1998; 
Nair et al., 1999). Furthermore, attempts has been made to use a single-point isotherm to 
characterize P retention in such soils (Mozaffari and Sims, 1993; Harris et al., 1996; Nair 
et al., 1998). A single point isotherm indexing approach, termed the relative phosphorus 
adsorption (RPA) index, effectively arrayed sandy Florida soil samples with respect to 
relative P adsorption (Harris et al., 1996). 

Although the equilibrium P concentration in the soil solution is generally 
relatively low, recent studies have shown that the P concentration in the soil solution can 
increase significantly well before the soil adsorption maximum has been reached 
(Breeuwsma and Silva, 1992). The Dutch have developed a test referred to as the 



63 



64 

"Degree of P Saturation" (DPS) which relates the soil P sorption capacity to an 

extractable P concentration as follows: 

DPS = Extractable soil P x 100 

P sorption maximum 

Operationally, DPS can be defined as oxalate-extractable P divided by the 
phosphate sorption capacity of the soil that is estimated from equations including oxalate- 
extractable Fe and Al (Breeuswma et al., 1995) as follows: 

DPS- Pox X 100 

Fe ox + Alox 

The Dutch used a reference soil solution concentration of 0. 1 mg P/L as a critical 
concentration based on water quality studies. They found that a DPS value of 25% would 
generally result in soil solution concentration equal to greater than 0. 1 mg P/L. Pautler 
and Sims (1998) study on comparison of short-and long-term sorption kinetics in Atlantic 
coastal plain soils concluded that the potential for P loss from over-fertilized soils can be 
improved by a knowledge of DPS of soils. 

Sharpley (1995) found a single relationship (r^ of 0.86) to describe the 
concentration of dissolved phosphate (DP) as a function of P-sorption saturation for ten 
soils ranging from sandy loam to clay in texture. The Mehlich-3 extractant was used for 
extractable soil P and the Langmuir P-sorption maximum as P-sorption capacity in the 
calculation of DPS. 

This study was conducted to evaluate the P retention of a sandy soil under dairy 
effluent application using traditional muhipoint isotherms, RPA, and DPS. 



:-/ 



65 

Materials and Methods 
Experiment Location and Design 

The site of the study was located at North Florida Holstein Dairy facility, which is 

two miles south of Bell, Florida. A randomized block design containing three blocks and 
arranged as a split plot was used as the experimental design. Main plots were N loading 
rates and subplots were cropping systems. Subplot area was 232 m . Dairy waste effluent 
was used as the N source. The N application rates were 448 and 896 kg/ha/yr which 
correspond to P loading of 112 and 224 kg/ha/yr. The cropping systems were corn-forage 
sorghum-rye, and perennial peanut-rye. 
Soil Selection and Sampling 

The soil was mapped as a Kershaw sand (sandy, thermic, uncoated Typic 
Quartzipsamments). Soil profile samples (0-15, 15-30, 30-45, 45-60, 60-80, 80-100 cm) 
were collected in 1996 (prior to effluent application) and in 1997, and 1998 (after effluent 
application). Soil from three profiles in each subplot was collected, composited, mixed 
thoroughly and a 1-kg subsample was brought to the laboratory for analysis. Soil samples 
were air-dried and sieved (2mm) prior to analysis. Soil samples were also collected in a 
similar manner from an adjacent native area believed to unimpacted by manure or 
fertilization application. 
Soil Characterization 

Rapid chemical assessment of relative phosphorus adsorption (RPA) was done by 
procedure developed by Harris et al., (1996). Ten- gram samples of air-dry soil were 
weighed into 20-mL scintillation vial and 2 mL of a 2000 mg/L P solution was added. 
The content of the vials were mixed by vigorous shaking, and allowed to equilibrate for 
24 hours at room temperature. The contents was transferred from the vials to centriftige 



66 

tubes, and centrifuged at 1500 g for 5 min. The centrifuge tubes had small holes drilled 
through the bottom. During centrifugation, solution passed through the holes into small 
cups attached to the bottom of the centrifuge tubes. Solution was removed from the cups 
and passed through a 0.45-|im syringe filter. Phosphorus in the solution was determined 
by the method of Murphy and Riley (1962) at an absorbance at 880 nm. The relative P 
adsorption capacity was quantified by dividing the absolute amount of P adsorbed by the 
maximum possible under these conditions, which was 400 mg/kg. 

Phosphorus multipoint adsorption isotherms were measured using 2 g of an air- 
dried soil treated with 20 mL of O.OIMKCI solution containing various levels of P (0, 
0.2, 0.5,1, 5, 10, 40, and 100 mg/L) in 50-mL centrifuge tubes. The tubes were placed on 
a mechanical shaker for 24 hours equilibration period. At the end of 24 hours period, the 
soil was centrifuged at 3620 x g for 10 min. The supernatant was then filtered through a 
0A5-\xm membrane filter and the filtrate analyzed for P (Murphy and Riley, 1962). 

Total phosphorus (TP) was determined by ashing 1.0 g of soil for 3 hours and 
then solubilizing with 6MHC1 (Anderson, 1976). Double-acid (Mehlich I)-extractable P, 
Al, Fe, Ca and Mg were obtained with a 1:4 soil/double acid ratio (Mehlich, 1953). 
Phosphorus in solution was analyzed by the molybdenum-blue method (Murphy and 
Riley, 1962). Soil pH was determined on 1:2 soil/water ratio, and the organic carbon 
content of the air-dried samples was determined by combustion (Broadbent, 1965). 
Texture was determined using the pipette method (Day, 1965). 

Oxalate-extractable P, Al, and Fe were determined by extraction with an 
ammonium oxalate (0. 1 M oxalic acid + 0. 175 M ammonium oxalate) solution adjusted to 
pH 3.0 (McKeague and Day, 1966). The suspension was equilibrated for 4 hours with 



67 

continuous shaking, centrifuged, filtered through a 0.45-^im filter and analyzed for P, Al, 

and Fe. 

Calculations 

Degree of P saturation (DPS) was calculated as oxalate-extractable P divided by 

the P sorption capacity of the soil, which is estimated as the sum of oxalate-extractable Fe 

and Al (Breeuwsma et al., 1995). This DPS is referred to in this study as (DPS - 1). Also, 

DPS was calculated as double acid (Mehlich I)-extractable P divided by the P sorption 

capacity of the soil, estimated from the sum of oxalate-extractable Fe and Al. This DPS is 

referred to as (DPS - 2). 

Extractable soil P x 100 

DPS = P- sorption capacity 

P sorption capacity was estimated from oxalate-extractable Al and Fe. 
Adsorption parameters were calculated using the Langmuir adsorption equation: 

i * 

C/S=l/kS„ax+C/S„ax . ^ -^^ 

Where 

S = S' + So the total amount of P sorbed, mg/kg 
S' = P sorbed by the solid phase, mg/kg 
So = originally sorbed on the solid phase, mg/kg 
C = concentration of P after 24 h equilibration, mg/L 
Smax = P sorption maximum, mg/kg 
k = constant related to the bonding strength, L/mg P 
So was estimated using a least square fit of S' measured at low equilibrium 
concentration, C. At these concentrations, the linear relationship between S' and C can be 



68 

described by S' = K C - So where K is the linear adsorption coefficient (Graetz and Nair, 
1995). Po ( soluble P) referred to P in solution after a 24-h equilibrium period when no P 
was added. 

Equilibrium P concentration (EPC), was defined as the concentration of P in 
solution where adsorption equal desorption and was the value of C when S' = 0. 
Statistical Analysis 

Data analyses were done using SAS (SAS Institute Inc. 1985) program (PROC 
MIX) procedure (SAS Institute Inc. 1992). Relationships among parameters were 
evaluated using linear correlation. Multiple regression was used to examine the strength 
of the relationships between parameters. 

: i ' - .-• O 

Results and Discussion 
Relative Phosphorus Adsorption (RPA) 

After two years of effluent application, RPA values of soil samples from the study 
site did not show any significant change and remained in the same range reported before 
effluent application (Table 4-1). The absence of significant differences in RPA values pre 
and after effluent application, indicated that effluent application did not influence P 
sorption capacity for this soil which has been heavily loaded with animal manure prior to 
the start of the study. 

The RPA values of soil samples from the study site ranged from 0.5 to 0.6 
throughout the profile pre-and post effluent application (Table 4-1). The samples from 
native area adjacent to the study site showed an RPA value of 0.8 to 0.9 through out the 
soil profile (Table 4-1). However, part of this difference could be due to differences in 
clay content between native and study site samples as shown in Chapter 2 (Table 2-3). 



69 



Table 4-1. RPA values within the soil profile of the study site (n = 12 profiles) prior and 
after to application of effluent compared to the "native soil" (n = 1 profile). Values are 
Least Square Mean (LSM). 



Depth 


1996 


1997 


1998 


Native 


0-15 


0.47 


0.57 


0.58 


0.82 


15-30 


0.52 


0.58 


0.60 


0.93 


30-45 


0.54 


0.59 


0.61 


0.93 


45-60 


0.62 


0.60 


0.61 


0.93 


60-80 


0.63 


0.60 


0.57 


0.90 


80-100 


0.64 


0.63 


0.57 


0.89 



70 



In the study by Rhue et al, 1994, RPA for Quartzipsamment was correlated with clay 
content (R^ = 0.87). 

RPA = -10.076 + 128.769 log (clay + 1) 
Assuming that the relationship would apply to the soil at the study site and using the clay 
contents in (Table 2-3), the RPA for native soil should have been about 0.60 while that 
for the study site soil should have been about 0.45, showing the relative effect of clay 
content on RPA, Why these measured RPA values were higher than those predicted by 
the equation of Rhue et al., (1994) is not known. The effect of P loading on RPA has not 
yet been explored. The relative contribution of clay and P loading cannot be related from 
this data. The lower RPA values of the study site compared to the RPA values of the 
native site indicated lower relative P adsorption capacity of the study site. Single-point 
isotherm has been used to effectively index sandy materials. For example, Harris et al. 
(1996) stated that RPA effectively arrayed sandy Florida soil samples with respect to 
relative P adsorption. Their soil samples included five taxonomic groups for sandy 
surface and subsurface horizon groupings. The RPA values were 0.74 and 0.69 for A and 
Bt horizons of Paleudults, 0.54 for coated Quartzipsamments (defined as coated) in the 
surface horizon and 0.58 in the subsurface, 0.48-0.47 for uncoated Quartzipsamments 
(defined as slightly coated) surface and subsurface horizons, 0.26-0.08 for uncoated 
Quartzipsamments (defined as clean) surface and subsurface, and 0.05-0.01 for Alaquods 
surface and subsurface horizons, respecfively. They also pointed out that RPA does not 
directly provide values for maximum P adsorption, but it closely relates to such values 
derived fi-om P adsorption isotherms. Mozaffari and Sims (1994) also have evaluated 
single-point isotherms after Bache and Williams P sorption index (PSI) and indicated 



71 

that PSI may be a viable alternative to sorption isotherms for the purpose of a rapid 
means to assess the ability of a soil profile to retain additional P. The PSI in their case 
was found to be highly correlated (r^ = 0.94) with the Langmuir P sorption maxima 
except when PSI exceeded 1400 mg/kg, where significant non-linearity was observed. 
Recently, Nair et al., (1998) found that single point sorption values measured at 1000 mg 
P/kg for soils of the Bh and Bw horizons from a low manure-impacted (pasture) and a 
high manure impacted (holding) areas were comparable (r^ = 0.98) to the S^ax values 
calculated using a Langmuir equation and concluded that a single point sorption value 
was a very convenient and quick means of characterizing the soils for maximum P 
sorption capacity. 

In this study, values of RPA for soil samples fi-om the study site compared to the 
RPA values fi-om a native site indicated clearly a low sorption capacity for the soil 
samples from the study site throughout the profile to Im depth. The absence of 
significant differences in RPA values for different horizons within the profile could be 
due to the absence of differences in soil constituents known to be responsible for P 
sorption, such as clay content, between horizons within the profile. However, RPA values 
were correlated r^ = 0.65 with double acid (Mehlich I)-extractable P, and Al (Table 4-2). 
Similarly, RPA also correlated with oxalate-extractable P, Al, and Fe (r^ - 0.63, n = 72). 
Degree of Phosphorus Saturation (DPS^ 

Degree of phosphorus saturation (DPS - 1) values of soil samples in the study site 
varied with depth in the soil profile. DPS - 1 indicated a 50% saturation in surface 
horizon, 26% at the 30-45 cm depth, 13-17 % at the 45-60 cm depth, and about 10-13% 
at lower depths of 80 and 100 cm pre-and post effluent application (Table 4-3). However, 



72 



Table 4-2. Multiple regression equations relating RPA to a) Mehlich I (DA) Al, Fe and P, 

b) Oxalate Al, Fe, and P in 1996 (prior to the application of effluent) (n = 72). 

No. Equation Model 



R^ 



a) RPA = 0.314+ 0.0022DA-A1*** + 0.01181 DA-Fe 0.65*** 

- 0.00021 DA-P*** 



b) RPA = 0.515 + 0.0015 OX-Al***- 0.00167 OX-Fe** 0.63*** 

- 0.00087 OX-P*** 



Table 4-3. DPS - 1^ % values within the soil profile of the study site (n = 12 profiles) 
prior and after to application of effluent compared to the "native" soil (n = 1 profile). 
Values are Least Square Means (LSM). 



Depth, cm 1996 \991 1998 Native 

0-15 49.63 40.14 48.25 18.70 

15-30 42.46 41.00 41.52 15.30 

30-45 26.53 26.02 25.12 12.43 

45-60 13.30 21.95 17.89 10 94 

60-80 9.98 12.03 13.93 12 43 

80-100 9.07 10.24 1182 10 89 

t DPS - 1 % - (P„, / (Feox + Alox)) X 100 



73 

the soil samples from native area adjacent to the study site showed DPS - lvalue of about 
19% to 1 1% through out the soil profile (Table 4-3). 

The DPS - 2 values for soil samples of the study area ranged from about 37% at 
the surface to 4% at a depth of 100 cm compared to 9% to 5% for soil samples from the 
native area, respectively (Table 4-4). Values of DPS - 1 and DPS - 2 for soil samples of 
the study area were highly correlated in strong relationship (r^ =0.92, n = 144) (Fig. 4-1). 
These resuhs indicated that the surface horizon is more likely to release P than the deeper 
depths. Sharpley (1995) found that a P saturation of 25%, the critical value used in the 
Netherlands, would support a DP (Dissolved-P) concentration in surface runoff of 0.69 
mg/L using Mehlich-3. Phosphorus sorption saturation, in his study, was calculated from 
Mechlich-3 extractable-P and Langmuir P-sorption maximum. However, in Florida, 
Mehlich I is the common soil P-test and the use of a common STP to express the DPS 
might be practically useftil. If the DPS can be determined by a standard soil test 
procedure and commonly used as Mehlich I, the DPS can become a usefiil tool for 
evaluating and comparing areas of potential P losses. Also, the strong relationship 
between DPS - 1 and DPS - 2 suggested by this study, could be used to compare values 
of both DPS if this relationship is similar enough in other soils. 
Langmuir Adsorption Parameters 

Surface horizons from the study site prior to the application of effluent showed a 
lower Langmuir P- sorption maximum (55 mg/kg) associated with higher equilibrium P 
concentration (EPCo) and a higher P originally sorbed (So) compared to subjacent 
horizons (Table 4-5). These differences were significant (/'<0.01) for equilibrium P 



74 



Table 4-4. DPS - 2^ % values within the soil profile of the study site (n = 12 profiles) 
prior and after to application of effluent compared to the "native" soil (n = 1 profile). 

Values are Least Square Means (LSM). 

Depth, cm 1996 1998 Native 



0-15 36.08 37.07 8.79 

15-30 31.29 29.68 6.63 

30-45 10.98 15.59 4.85 

45-60 9.32 7.975 3.94 

60-80 5.57 5.57 4.85 

80-100 3.58 3.77 4.49 
t DPS - 2 % = ( Mehlich I extractable-P / (Feox + Alox)) x 100 



75 



90 



y = 1.05441 + 6.0035 
R^ = 0.9214 




10 20 30 40 50 

DPS -2 



60 



70 80 



Figure 4-1. Relationship between Degree of P saturation (DPS - 1) calculated from 
oxalate extractable-P and Degree of P Saturation calculated from Mehlich I (DPS - 2) for 
soil samples from the study site. 



76 



Table 4-5. Comparison of Langmuir parameters (Smax, EPCo, k) and So mean values of 
different horizons within the soil profile prior to the application of effluent in 1996 and 

after two years of effluent a pplication in 1998. 

Year Horizon Smax So EPCo k 

mg/kg mg/kg mg/L L/mg 



1996 


A 


55 


28.39a* 


8.81a 


0.15 




0-1 5cm 


(13.2-76.3) 


(14.9-39.6) 


(2.31-14.89) 


(0.04-0.40) 




CI 


100 


5.16a 


1.19b 


0.37 




30-45cm 


(72.5-154) 


(1.1-8.8) 


(0.05-3.87) 


(.048-0.64) 




C2 


95 


4.23b 


0.79b 


0.54 




45-60 cm 


(37-142.8) 


(0-4.23) 


(0-0.79) 


(0.2-0.99) 


1998 


A 


88 


25.63a 


5.02a 


0.12 




0-15 cm 


(70.5-97.1) 


(18.11-31.4) 


(3.85-6.18) 


(0,04-0.42) 




CI 


95 


10.05b 


1.64b 


0.68 




30-45 cm 


(27.7-153) 


(2.29-25.8) 


(0.5-4.0) 


(0.07-2.8) 




C2 


96 


4.67b 


0.57b 


0.74 




45-60 cm 


(18.9-175) 


(2.1-8.34) 


(0.04-2.0) 


(0.09-2.44) 



* LS mean values for given parameters followed by the same latter are not significantly 
different (p<0.0\). Numbers in parentheses are the highest and lowest value for the 
parameter (n= 18). 






77 

concentration (EPCo) and P originally sorbed (So). There were no significant difference in 
Langmuir parameters between 1996 and 1998. The same trend of a lower Langmuir P- 
sorption maximum associated with higher equilibrium P concentration (EPCo) and a 
higher P originally sorbed (So) continued in 1998 after the application of effluent. The 
absence of differences in Langmuir parameters at the beginning and end of the study 
period could be attributed to the variability usually associated with such measurements, 
the study time limitation, and the fact that this site was heavily loaded with animal 
manure prior to the start of the study. However, equilibrium P concentration (EPCo) 
showed a strong relationship (r^ = 0.94) with DPS - 1 (Fig. 4-2). Based on this 
relationship, a DPS - 1 value of 20 % corresponds to an EPCo value of approximately 1 
mg/L. Another parameter, from the isotherm study Po (P in solution after a 24-h 
equilibrium period when no P was added (soluble P)), also showed a strong relationship 
(r^ =0.92) with DPS - l(Fig. 4-3). Based on this relationship, a DPS - 1 value of 20% 
corresponds to a Po of approximately 5 mg/L. Such correlation between DPS and 
Langmuir parameters suggests that an integration of such tools could be used in the study 
oftheassessment of the tendency of this soil to release P. t' ... 

Summary and Conclusions 
This study demonstrated the possibility of integrating a numbers of tools to 
characterize soil P retention at the study site. The use of a single point isotherm such as 
relative P adsorption (RPA) showed that the soil at the site has a lower relative adsorption 
for P compared to soils samples from a native site. After two years of effluent 
application, RPA values of soil samples from the study site did not show any significant 
change and remained in the same range reported before effluent application. The absence 



78 



o 

Q. 
UJ 



y = 0.161x- 1.9555 
1^^ = 0.9408 




ri ,.. i :c 



so 



60 



Figure 4-2. Relationship between Degree of P saturation calculated from oxalate 
extractable-P (DPS - 1) and equilibrium P concentration (EPCo) for soil samples from the 
study site. 



79 



25 



20 



15 



D) 

E 
o 
a. 



10 



y=0.5078x- 4.5994 
R^= 0.9257 




50 



60 



Figure 4-3. Relationship between Degree of P saturation calculated from oxalate 
extractable-P (DPS - 1) and soluble P (Po) mg/L for soil samples from the study site. 



80 

of differences in RPA values pre-and post effluent application, indicated that effluent 
application did not influence P sorption capacity for this soil which has been heavily 
loaded with animal manure prior to the start of the study. 

The degree of phosphorus saturation (DPS) showed that soil samples from the 
study site were 50% saturated at the surface compared to about 19% for the surface soil 
samples from the native site. These results indicated that the surface horizon is more 
likely to release P than the deeper horizons. 

The isotherm study for this soil was also in agreement with the above finding 
where surface horizon showed a lower Langmuir P- sorption maximum (55 mg/kg) 
associated with higher equilibrium P concentration (EPCo) and a higher P originally 
sorbed at the solid phase (So) compared to subjacent horizons. The same trend of a lower 
langmuir P- sorption maximum associated with higher equilibrium P concentration 
(EPCo) and a higher P originally sorbed at the solid phase (So) continued in 1998 after the 
application of effluent, however no significant changes were observed between the two 
years. 

Values of degree of P saturation (DPS - 1) and (DPS - 2) were highly correlated 
(r =0.92), which suggest the possibility of integrating the most common STP in the 
region (Mehlich I) into the useful approach of degree of P saturation. Also, DPS - 1 was 
highly correlated with equilibrium P concentration (EPCo) (r^ = 0.94), and with soluble P 
(Po) (r =0.92). However, further research is needed to determine whether these 
relationships are similar enough in other sandy soils to be valuable as a tool in predicting 
the tendency of soil to release P. 



CHAPTERS 

DOWNWARD PHOSPHORUS MOVEMENT ASSESSMENT IN A SANDY SOIL 

RECEIVING DAIRY WASTE EFFLUENT 



Introduction 
Loss of P from land can occur in three ways; as water-soluble and/or particulate P 
in surface runoff, as water-soluble and/or particulate P in subsurface runoff (leaching), 
and as water-soluble and/or particulate in flow to groundwater, referring to P picked up 
by water that passes to the water-table and which is subsequently discharged to streams, 
rivers or lakes as seepage (Ryden et al., 1973). P leaching has normally been considered 
to be inconsequential in most soils, but recent studies have found that there are a 
combination of agriculture management practices, soil properties, and climatic conditions 
that can result in significant P accumulation in subsoils. Whether or not P that leaches 
into subsurface horizons is later transported to water bodies depends on the depth of 
leaching and the hydrological connections of the watershed (Sims et al., 1998). The 
association of P accumulation with its downward movement has been the subject of 
numerous studies in soils amended with commercial fertilizers and /or organic wastes. 
Studies by King et al. (1990), Kingery et al. (1994), Mozafifari and Sims (1994), and 
Eghball et al. (1996) reported P leaching to -75 cm depending on factors such as soil 
type and the amount of P accumulated in the surface horizon. Furthermore, Eghball et al. 
(1996) suggested a greater downward mobility for organic forms of P. Previous studies 
from Florida also illustrated the extent of P leaching that can occur in deep, sandy soils. 
One of the earliest studies in Florida was that of Bryan (1933) who reported P leaching to 



-I ■ 



81 



-":-V|(j;nrT^-A— , 



82 

depths of at least 90 cm in heavily fertilized citrus groves of varying ages. Humphreys 
and Pritchett (1971), in their study of six soil series in northern Florida, 6 to 10 years 
after applying superphosphate, reported extensive P leaching and subsequent 
accumulation in the spodic horizon of a Leon fine sand. They noted that all fertilizer P 
had leached below a depth of 50 cm in the Pomello and Myakka soil series. A study by 
Wang et al. (1994) found that high levels of P could be leached from surface (Ap) 
horizons of four sandy Florida soils heavily loaded with dairy manure despite high pH 
and abundant Ca^"^ in solid and solution phases. Graetz and Nair (1995), Nair et al. 
(1995), Nair et al. (1998), and Nair et al. (1999), in a series of studies on Spodosols in the 
Lake Okeechobee basin of Florida, concluded that the P that leaves the surface (A) 
horizon might be lost through surface and subsurface drainage. The P portion that reaches 
the spodic (Bh) horizon will be held as Al-and Fe-associated P, either in the inorganic or 
in the organic fraction. The high percentage of HCl-extractable P (Ca-and Mg-associated 
P) in the A horizon of the intensive dairy component was also of potential concern. The 
HCl extractable P could be continuously extracted by NH4 CI or by water (Graetz and 
Nair, 1995), suggesting that about 80% of the total soil P had the potential to move 
eventually with drainage water into Lake Okeechobee. Recently, Sims et al. (1998) 
reviewed current research on P leaching and loss in subsurface runoff in Delaware, 
Indiana, and Quebec. They concluded that the situation most commonly associated with 
extensive P leaching, and thus the increased potential for P loss via subsurface runoff, has 
been the long-term use of animal manures. 

The most common soil P tests used to assess P status are the traditional agronomic 
soil tests for P such as Mehlich I, Mehlich 3, Bray I, and Olsen. These tests are often well 



83 

correlated with environmentally oriented P tests such as biologically available P (BAP) 
and dissolved reactive P (DRP) in runoff (Pote et al., 1996). Water soluble P (WSP) in 
particular has been characterized as an appropriate environmental soil P test (Sharpley et 
al., 1996, Moore et al., 1998). Therefore, this study was initiated to assess the vertical 
movement of P in the soil profile during application of dairy waste effluent to two 
cropping sequences in a deep sandy soil, using WSP and labile P concentrations within 
the profile as indicators of downward P movement. 

Materials and Methods 
Experiment Location and Design 

The study site was located at North Florida Holstein Dairy facility, which is two 
miles south of Bell, Florida. A randomized block design containing three blocks and 
arranged as a split plot was used as the experimental design. Main plots were N loading 
rates and subplots were cropping systems. Subplot area was 232 m^. Dairy waste effluent 
was used as the N source. The N application rates were 448 and 896 kg/ha/yr which 
correspond to P loading of 112 and 224 kg/ha/yr. The cropping systems were corn-forage 
sorghum-rye, and perennial peanut-rye. 
Soil Selection and Sampling 

The soil was mapped as Kershaw sand (sandy, thermic, uncoated Typic 
Quartzipsamments). Soil profile samples (0-15, 15-30, 30-45, 45-60, 60-80, 80-100 cm) 
were collected in 1996 (prior to effluent application) and in 1997, and 1998 (after effluent 
application). Soil fi-om three profiles in each subplot was collected, composited, mixed 
thoroughly and a 1-kg subsample was brought to the laboratory for analysis. Soil samples 
were air-dried and sieved (2mm) prior to analysis. Soil samples were also collected in a 



84 . 

similar manner from an adjacent native area believed to unimpacted by manure or 
fertilization application. 
Soil Characterization 

Texture was determined using the pipette method (Day 1965). Total phosphorus 
(TP) was determined by ashing 1.0 g of soil for 3 hours and then solubilizing with 6 A/ 
HCl (Anderson, 1976). Double-acid (Mehlich I)-extractable P, Al, Fe, Ca and Mg were 
obtained with a 1:4 soil/double acid ratio (Mehlich, 1953). Phosphorus (P) in solution 
was analyzed by the molybdenum-blue method (Murphy and Riley, 1962). Soil pH was 
determined on 1:2 soil/water ratio, and the organic carbon content of the air-dried 
samples was determined by combustion (Broadbent, 1965). 

Water soluble P was extracted using a 1: 10 (soil: 0.01 A/Ca Cb) ratio, by 
shaking the sample end- over-end for 1 hour, centrifuging for 20 min (1000 g), and 
filtering (0.45 um). Phosphorus in solution was analyz;ed by the molybdenum-blue 
method (Murphy and Riley, 1962). Labile P was obtained using a fractionation scheme. 
The scheme used to fractionate soil-P was a modification by Nair et al. (1995) of that of 
Hieltjes and Lijklema (1980). A 1-g air- dried sample was extracted twice with 25 mL of 
1 MNH4CI (adjusted to pH 7.0) (two hours shaking). After each extraction, the content 
were centrifiaged for 15 min at 3620 x g and filtered through a 0.45-|im filter. All 
extractions were carried out at room temperature. P determination was done using the 
procedure of Murphy and Riley (1962) on a spectrophotometer at wavelength of 880 nm. 
NILiCl-extractable P was defined as labile P (Petterson and Istvanovics, 1988). 
Statistical Analysis 

Data analyses were done using SAS program (PROC MIXED) procedure (SAS 
Institute Inc. 1985). Relationships among parameters were evaluated using linear 



85 ; , 

correlation. Multiple regression was used to examine the strength of the relationships 
between parameters. 

Results and Discussion 

The soil from the study site showed a higher content of WSP prior to the 
application of effluent in 1996 compared to the WSP in the soil profile of the native soil 
collected from an adjacent site. WSP for soil samples from the study site ranged from 
19.6 mg/kg at the surface to 1.9 mg/kg at 100 cm compared to 0.4 mg/kg at the surface 
and <0.1 mg/kg at lower depths of the native soil (Table 5-1). This higher content of 
WSP in the soil samples of the study site was associated with higher Mehlich I- 
extractable P as shown in (Table 5-2). 

The effect of date*rate (P< 0.001), crop*rate (P< 0.001), and date*depth (P< 
0.001) were significant for WSP. The application of effluent caused an increase in WSP 
content at all depths except the surface soil in 1 998 and the change in WSP content was 
significant for both rates (Table 5-1). The high effluent application rate showed a 
decrease in the WSP content at the surface to about 30 cm then an increase down to the 
100-cm depth (Table 5-3, Fig. 5-1). Similarly, WSP concentrations for the low effluent 
application rate increased at lower soil depths (Table 5-4, Fig. 5-2). However, though the 
trend of change in WSP content under the high and low rates effluent application at the 
surface horizon were similar, the trend of change in WSP under low rate application at 
the depths of 15-30, 30-45, and 45-60 cm was unexplainable. It was expected that such an 
increase in WSP concentrations at these soil depths would be acceptable for the high 
effluent rate application. Nevertheless, when WSP averages in all depths within the soil 



86 



Table 5-1. Mean water soluble P concentrations (WSP) in the soil profile prior to 
application of effluent (1996) and after application (1998) (n = 12 profiles) compared to 
native soil (n = 3 profiles). Values are least square means (LSM). 



Depth 1996 1998 Native 

(cm) mg/kg 

0-15 19.6 13.9 0.4 

15-30 16.2 22.2 0.1 

30-45 9.10 23.2 <0.1 

45-60 4.30 13.1 . <0.1 

60-80 2.80 8.10 <0.1 

80-100 1.90 5.40 <0.1 



i^* 9 



87 



Table 5-2. Mehlich I-extractable elements concentrations and total P (TP) in "native" soil 
(n= 1 profile) a.n d study site soil profi les (n = 

Location Depth Ca Mg Al Fe P TP 
(£!B) "^g^g 

Native 0-15 11.7 1.9 267 18.4 47 214 



15-30 


5.1 


1.1 


317 


20.7 


52 


270 


30-45 


6.0 


0.8 


330 


19.1 


39 


241 


45-60 


4.7 


0.7 


337 


16.3 


36 


184 


60-80 


4.7 


0.8 


308 


16.3 


39 


181 


80-100 


4.1 


0.7 


280 


14.3 


33 


173 


StudvSite 0-15 


968 


115 


301 


23.5 


283 


328 


15-30 


522 


69.3 


280 


22.8 


184 


254 


30-45 


208 


34.1 


203 


19.6 


75 


154 


45-60 


135 


25.7 


161 


17.9 


37 


254 


60-80 


103 


22.9 


133 


15.9 


20 


218 


80-100 


75 


19.2 


117 


14.5 


12 


192 



-f ':■ ■ ' : '■% 









88 




1996 



■1998 



Figure 5-1. Mean water soluble P (WSP) concentrations within the soil profile of the 
study site under the high rate effluent application prior to the application of effluent in 
1996 and after effluent application in 1998. Values are LSM+ Std. Error. 



89 




120 J 



■1996 



■1998 



Figure 5-2. Mean water soluble P (WSP) concentrations within the soil profile of the 
study site under the low rate effluent application prior to the application of effluent in 
1996 and after effluent application in 1998. Values are LSM± Std. Error. 



;W. 



90 



Table 5-3. Changes in WSP concentration within the soil profile under high application 
rate after the application of effluent (1998) vs. prior to the application of effluent (1996). 
Depth (cm) 1996 vs. 1998 

0-15 -9.78** 

15-30 NS 

30-45 NS 

45-60 NS 

60-80 NS 

80-100 NS 

** Significant at the 0.001 probability levels, NS = none significant. 



Table. 5-4. Changes in WSP concentration within the soil profile under the low 
application rate after the application of effluent (1998) vs. prior to the application of 
effluent (1996). 
Depth (cm) 1996 vsj 998 

0-15 NS ' 

15-30 13.8** 

30-45 23.4** 

45-60 10.8** 

60-80 5.3** 

80-100 NS 



** 



Significant at the 0.001 probability levels, NS = none significant. 



91 i.^\i W'> 

profile in 1998 was compared to those of 1996 (Fig. 5-3), it showed a trend similar to that 
found for Mehlich I-extractable P (Chapter 2). The trend may mimic vertical P 
movement in the soil profile and suggested that the decrease in the surface horizon may 
be attributed to both crop uptake of P and the leaching effect of effluent irrigation. 

The labile P fraction or easily removable P as defined by (Petterson and 
Istvanovics, 1988) behaved similarly to WSP regarding its increase in the lower depth of 
the profile. As mentioned in Chapter 3, a comparison of labile P mean concentrations in 
each depth within the soil profile at the beginning (1996) and end of the study period 
(1998) showed that the increase was statistically significant (P < 0.001- P < 0.05) at all 
depths. Labile-P constituted 18-40% of TP in the soil profile. Labile-P in 1996 ranged 
fi-om 85 in the surface horizon to 37 mg/kg at the lower depth (100 cm) which 
corresponded to 26 and 18% of TP, respectively. The application of effluent increased 
this fraction to 97 mg/kg in the surface horizon and 69 mg/kg at the lower depth (100 cm) 
inl998. Labile-P increased at the surface horizon and throughout the profile over time 
with substantial increases in the lower depth accounting for 40% of TP in 1998 (Fig. 5-4). 
Previous research has shown that traditional agronomic soil P tests are often correlated 
with dissolved P and /or bioavailable P in runoff waters and subsurface drainage. Wolf et 
al. (1985) reported that the equilibrium P concentrafion at zero sorption (EPCo) and algal- 
available P (extracted by a . IMNaOH + IN NaCl solution) could be accurately 
predicted in a wide range of U.S. soils by the Bray Pi, Mehlich 1, and Olsen soil P tests. 
Pote et al. (1996) also reported that water soluble P (WSP) was well correlated with 
runoff P in a field study with tall fescue. 



92 



Q. 



60 



80^ 



-we- 



^2^ 




25 



WSP 1996 



-WSP 1998 



Figure 5-3. Mean water soluble P (WSP) concentration within the soil profile of the study 
site prior to the application of effluent in 1996 and after effluent application in 1998. 
Values are LSM± Std. Error. 



V 






93 




1996 -♦-1998 



Figure 5-4. Labile-P concentration within the soil profile of the study site prior to the 
application of effluent in 1996 and after effluent application in 1998. Values are LSM+ 
Std. Error. 

■'■■■■rim.€ 



'V\^ . 



94 

In this study, a number of correlations were investigated to determine the 
strongest relationship between parameters. Labile P correlated with Mehlich I-P (r^ = 
0.84, P<0.001) and WSP correlated with Mehlich I-P (r^ = 0.49, P<0.001) in 1996. The 
low r^ value for WSP is due to high variability in WSP data. The significant relationships 
(r^ = 0.84) between Mehlich I-P and labile P and Mehlich I-P and WSP (r^ = 0.49) 
showed that Mehlich I-P is a good indicator of leachable P and/or P in the runoff Pote et 
al. (1996) reported a significant relationship (r^ = 0.82, P<0.001) between WSP in surface 
soil and dissolved reactive P (DRP) in runoff and between Mehlich-3 in surface soil and 
dissolved reactive P(DRP) in runoflf(r^ = 0.72, P<0.001). v. 

The results of this study also showed a significant relationship between labile P 
and degree of phosphorus saturation. Labile P correlated with DPS - 2 (r^ = 0.71, 
P<0.001) and DPS - 1 (r^ = 0.62, P<0.001) in 1996. In contrast, WSP did not correlate as 
well as labile P with DPS - 1 (r^ - 0.57, P<0.001) and DPS - 2 (r^ = 0.49, P<0.001). 
These significant relationships between labile P and DPS and WSP and DPS showed the 
link between easily removable P and degree of P saturation. Pote et al. (1996) reported a 
significant relationship (r^ = 0.75, P<0.001) between DPS - 1 and dissolved reactive P 
(DRP) in runoff The link between P concentration in soil solution and the degree of P 
saturation was suggested by Breeuwsma and Silva (1992), and results from the study site 
agree with this conclusion. 

The correlation between labile P and DPS - 1 held true after two years of effluent 
application (r^= 0.76, P<0.001), however in the case of labile P and DPS - 2, the 
correlation coefficient decreased to 0.23. In the case of WSP, there was also a decrease in 
correlation coefficient for DPS-1 (r^ = 0.13) and DPS-2 (r^ = 0.22). This trend was also 



95 

noted for the relationship between Mehlich I-P, labile P, and WSP. The correlation 
coefficients between labile P and Mehlich I-P was r^ = 0.3 1 and between WSP and 
Mehlich I-P r = 0. 12. The results of the significant linear relationship between soil test P 
(Mehlich I), labile P and WSP, and labile P and WSP and DPS-1 and DPS-2 under the 
conditions of this study could be useful for future comparison with similar results of other 
soils. Downward (vertical) movement of P in this soil was suggested by both labile P and 
WSP data. During the two years of effluent application, both parameters showed either a 
decrease or a non significant change at the surface horizon, but a significant increase at 
the lower depths. ' 

Summary and Conclusions 

The association of P accumulation and downward movement has been the subject 
of numerous studies in soils amended with commercial fertilizers and/or organic wastes. 
Sims et al. (1998) indicated that the most common agricultural situation associated with 
significant downward movement of P has been the accumulation of P to "very high" or 
"excessive" levels in soils from continuous application of organic wastes. 

Phosphorus leaching has normally been considered to be inconsequential in most 
soils but recent studies find that there are combinations of agriculture management 
practices, soil properties, and climatic conditions that can result in significant P 
accumulation in subsoils. Downward P movement in this soil was suggested by both 
labile P and WSP data. During the two years of effluent application, both parameters 
showed a significant increase at the lower depths of the soil profile. Phosphorus that 
leaches into subsurface horizons is later transported to water bodies depending on the 
depth of leaching and the hydrological connections of the watershed. Labile P increased 



96 • 



n ) ^^i^ 



form 37 mg/kg at the start of the study to 69 mg/kg at 100 cm by the end of the study in 
1998. Water soluble P showed a decrease at the surface horizon but increased in the 
lower depths of the soil profile by the end of the study period. 

The results of the significant linear relationship between Mehlich I-extractable P 
and labile P (r^ = 0.84), and Mehlich I-extractable P and WSP (r^ = 0.49) 1996 could be 
useful for future comparison with similar results of other soils. If this relationship 
between soil test P (Mehlich I) and easily removable P proven to be valid for similar 
soils, it could be helpful in relating soil test P (Mehlich I) levels in soils to P movement 
within the soil profile. 

Labile P and WSP also showed a significant relationships with DPS - 1 and DPS 
- 2 1996 which suggests a link between P concentration in soil solution and DPS reported 
by Breeuwsma and Silva (1992). The correlation between labile P and DPS - 1 held true 
(r = 0.76) after two years of effluent application , but the other correlations decreased 
after two years of effluent application . 



CHAPTER 6 
UTILIZATION OF DAIRY WASTE EFFLUENT THROUGH SEQUENTIAL 

CROPPING 



Introduction 

Developing manure utilization plans that are agronomical ly, economically, and 
environmentally sound is a challenge. Issues like accelerated eutrophication, P or N 
limitation, transport mechanisms, source management, soil P level, environmental soil 
testing for P, and manure management have to be considered. Animal manure can be a 
valuable resource if it can be integrated into cost effective best management practices. 
However, the need for such plan supported by research, especially in areas of intensified 
dairy production and deep, sandy soil is urgent. Many factors may be involved in 
developing an environmentally sound plan for manure utilization management. 

Uptake of nutrients by agronomic crop sequenced over time is an effective, 
economical, and environmentally sound means of nutrient recovery especially if the 
cropping system meets the environmental demand. The environmental demand can be 
meet by maximizing nutrient uptake by the crops while meeting the need of dairy 
producers. Sweeten et al. (1995) reported that irrigation with dairy lagoon effluent 
enhanced forage quality yield and did not impair quality of runoff or vadose zone 
percolate under the conditions tested for two complete cropping years. Their cropping 
systems were summer-only coastal bermudagrass and a summer-winter coastal/wheat 
rotation. Land application of lagoon effluent at rates that were at or below soil test 
recommendations for total or available nitrogen resuhed in runoff quality and vadose 



97 



98 

zone percolate quality that were 94-99% lower in volatile solids, COD, N and P than 
concentration in the applied lagoon effluent. 

A research report of two years of study on the use of dairy manure effluent in a 
rhizoma (perennial) peanut based cropping system (French et. al. 1995) suggested that if 
N pollution is the major concern in a particular area, then the PP-R (year-round perennial 
peanut and rye) would be a good choice since it performed as well or better than the C- 
FS-R (com, forage sorghum, and winter rye) and C-PP-R (com planted into a perennial 
peanut sod, perennial peanut, and rye) systems. However, if P were the major concern, 
the C-FS-R and C-PP-R systems would be better choices. The C-FS-R and C-PP-R 
systems were superior to the PP-R rotation in P removal values. Though P concentration 
level in perennial peanut forage were generally higher than those in com and forage 
sorghum, they were not high enough to compensate for the much lower annual dry matter 
yield of the perennial peanut system. 

This study was initiated to evaluate the effectiveness of the cropping systems, 
corn-forage sorghum-rye and perennial peanut-rye in P removal under two dairy waste 
effluent application rates. 

Materials and Methods 
Experiment Location and Design 

The study site was located at the North Florida Holstein Dairy facility, which is 
two miles south of Bell, Florida. A randomized block design containing three blocks and 
arranged as a split plot was used as the experimental design. Main plots were N loading 
rates and subplots were cropping systems. Subplot area was 232 m^. Dairy waste effluent 



99 

was used as the N source. The N appHcation rates were 448 and 896 kg/ha/yr, which 
correspond to P loading of 1 12 and 224 kg/ha/yr. 
Sampling and Analysis 

Two cropping systems (corn-forage sorghum-rye and perennial peanut-rye) and 
two N application rates (448 and 896 Kg/ha/yr) which correspond to P loading of (1 12 
and 224 Kg/ha/yr) respectively were sampled. The various crops were harvested at the 
appropriate times. In the corn-forage sorghum-rye (C-FS-R) system, com was no-till 
planted into rye stubble and harvested in July. Forage sorghum was then no-till planted 
into existing com stubble. Following sorghum harvest, rye was planted for the winter 
season using a no-till grain drill. For the perennial peanut-rye (P-R) system, the perennial 
peanut was harvested three times during the warm- growing season. Rye was overseeded 
into the peanut sod in late fall for the cool season crop. 

Within each plot, a 9.3 m^ portion was harvested, weighed and subsampled. 
Ground forage subsamples were sent to the Forage Evaluation Support Laboratory 
(FESL) at the University of Florida, Gainesville for analysis. Parameters measured 
include dry matter yield, N, and P concentration. N and P analysis involved a 
modification of the standard Kjeldahl procedure (Gallaher et al., 1975), followed by 
automated colorimetry (Hambleton, 1977) using a Technicon Auto Analyzer. 

The data and the statistical analysis were provided by Woodard et al. (2000) and 
are used herein to relate crop uptake to P accumulation in the soil. Responses were 
analyzed by fitting mixed effect models using the PROC MIXED procedure of SAS 
(SAS Institute Inc., 1992) and years were considered as repeated measures. 



100 

Results and discussion 

Mean P removal in the 1996-97 and the 1997-98 seasons from the corn-forage 
sorghum-rye (C-FS-R) cropping system was significantly higher compared to the 
perennial peanut-rye cropping system (Fig. 6-1 and 6-2). Mean P removal in the 1996-97 
season from the corn-forage sorghum-rye (C-FS-R) cropping system was the highest 67.2 
kg/ha (Table 6-1) compared to the perennial peanut-rye cropping system (Table 6-2). 
Mean P removal in (1997-98) season for the corn-forage sorghum-rye (C-FS-R) cropping 
system accounted for about 62 kg P/ha (Table 6-1) with no significant difference in P 
removal between 1996-97 and 1997-98 seasons. However, the perennial peanut-rye 
cropping system removed only about 35-39 kg P/ha in (1996-97) season and 45-54 kg/ha 
in (1997-98) season (Table 6-2). The higher P removal of the corn-forage sorghum-rye 
cropping system than perennial peanut-rye was reflected in the soil data reported in 
Chapter 2. Mehlich I-extractable P concentrations in soil from perennial peanut-iye 
cropping system plots was higher than those of the corn-forage sorghum-rye. This 
difference in Mehlich I-extractable P concentrations was significant (P < 0.05) under the 
high application rate. 

The difference in dry matter yield between the cropping systems was significant 
in both seasons (Wooodard et el., 2000). Dry matter yield of the corn -forage sorghum- 
rye cropping system was 27 Mg/ha in both seasons (Table 6-3), while the perennial 
peanut-rye cropping system dry matter yield was 12 Mg/ha in 1996-97 season and 18 
Mg/ha in the 1997-98 season (Table-6-4). ■" T ' ■ ". : ' ' 

The effect of effluent application rate/year on P removal of the corn-forage 
sorghum-rye was not significant for both seasons, while the effect of effluent application 



101 



Table 6-1. P removed (kg/ha) by the corn-forage sorghum-rye cropping system under 
high and low application rates during the 1996-97 and 1997-98 seasons. (Data obtained 
from Woodard et al. 2000). 



Application rate 


Corn 


Forage 
sorghum 


Rye 


Total 






kg/ha 










1996-97 






High 


30.2 


17.9 


16.8 


65 


Low 


32.5 


19.0 
1997-98 


15.7 


67 


High 


31.4 


17.9 


14.6 


64 


Low 


31.4 


19.0 


10.0 


60 



102 



80 



70 



60 



Blow rate 1996 



lew rate 1996 



Hgh rate 1996 B high rate 1996 



com 



sorghum 




Crop 




total 



Figure 6-1. P removal (kg/ha) of corn-forage sorghum-rye during the 1996-97 and 1997- 
98 seasons. 



103 



Table 6-2. P removed (kg/ha) by the perennial peanut-rye cropping system under high 
and low application rates during the 1996-97 and 1997-98 seasons. (Data obtained from 
Woodard et al. 2000). 



Application 


P. peanut 


Rye 


Total 


rate 






kg/ha 










1996-97 






i! 


r^nd 


3rd 




High 


9 


8 


18 


35 


Low 


8 


8 


23 
1997-98 


39 




l! 


;^nd 


3rd 




High 


13 


15 


07 19 


56 


Low 


13 


15 


07 10 


54 



104 



60 



llowratel996 0higiiratel996 




Cro 



Figure 6-2. P removal (kg/ha) of perennial peanut-rye during the 1996-97 and 1997-98 
seasons. 



105 

rate/year on P removal of the perennial peanut-rye was significant in both seasons 
(Wooodard et el., 2000). The average removal of P for forage crops in this study, 
including perennial peanut in the second season, were in agreement with the reported P 
removal for such crops by French et al. (1995), and published book value (NRCS Manure 
Master, 1999). However, despite the higher P removal by the C-FS-R than the P-R 
cropping system, P removal by the cropping system did not alter the high level of the soil 
P that was already present before application of the effluent (Chapter 2). After a high 
level of soil test P has been attained, considerable time is required for significant P 
depletion as reported by McCoUum (1991). 

Summary and Conclusions 
Uptake of nutrients by agronomic crops sequenced over time is an effective, 
economical, and environmentally sound means of nutrient recovery. Cropping systems 
are needed to maximize nutrient uptake while meeting the needs of dairy producers. In 
this study, higher P removal was recorded for the corn-forage sorghum-rye in both 
seasons. The higher P concentration in dry matter of perennial peanut and rye was not 
high enough to offset the lower dry matter yield. The perennial peanut-rye cropping 
system removed less P than corn-forage sorghum-rye cropping system. However, despite 
the higher P removal by C-FS-R than the P-R cropping system, P removal by the 
cropping system did not alter the high level of soil P that was already present before 
application of the effluent. After a high level of soil test P have been attained, 
considerable time is required for significant P depletion. Further investigation is needed 
to determine the best application rate based on N or P after taking in consideration the 



106 



Table 6-3. Average dry matter yield of the corn-forage sorghum-rye during the 1996-97 
and 1997-98 seasons. 



Crop Dry matter yield (Mg/ha) 



1996-97 1997-98 

Com 15 12 

Forage 08 11 
sorghum 

Rye 04 04 

Total 27 27 



Table 6-4. Average dry matter yield of the perennial peanut-rye during the 1996-97 and 
1997-98 seasons. 



Crop 


Dry 


matter yield (Mg/ha) 








1996-97 




1997-98 




P. Peanut 


07 




13 




Rye 


05 




05 




Total 


12 




18 





107 

soil P that was already present in the soil before application, and the needs of dairy 
producers. 



CHAPTER? 
SUMMARY AND CONCLUSIONS 

The study site soil was mapped as Kershaw sand (sandy, thermic, uncoated Typic 
Quartzipsamments) in the Gilchrist County soil survey report (Soil Survey Staff, Gilchrist 
County, Florida, 1973). Since the publication of the report, the criterion for coated vs. 
uncoated family placement has been changed for the USDA soil taxonomic system (Soil 
Survey Staff, 1999). The sandy materials sampled in this study would meet the criterion 
for coated family (5 percent silt plus 2 times the clay content), based on the particle size 
analysis Also, some auger borings to 2 m revealed spodic horizons which indicated 
inclusions of Spodosols, and dark colors in the surface horizon in some areas qualify it to 
be an Umberic epipedon, which would result in classification as an Inceptisol (Umbrept) 
rather than a Psamment. Nevertheless, the soil was consistently sandy and similar to 
Kershaw with respect to use and management. 

Loss of P from land can occur in three ways; as water-soluble and/or particulate P 
in surface runoff, as water-soluble and/or particulate P in subsurface runoff (leaching), 
and as water-soluble and/or particulate P in flow to groundwater, referring to P picked up 

by water that passes to the water-table and which is subsequently discharged to stream, 

f ■ 

rivers or lakes as seepage (Ryden et al., 1973). P leaching has normally been considered 
to be inconsequential in most soils, but recent studies find that there are combinations of 
agriculture management practices, soil properties, and climatic conditions that can result 
in significant P accumulation in subsoils. Whether or not P that leaches into subsurface 
horizons is later transported to water bodies depends on the depth of leaching and the 

108 



109 

hydrological connections of the watershed (Sims et al., 1998). However, the most 
common agricultural situation associated with significant downward movement of P has 
been the accumulation of P to "very high" or "excessive" levels in soils from continuous 
application of organic wastes (manure, litter, and municipal or industrial wastes and 
waste waters) (Sims et al., 1998). The trend of P accumulation and leaching has also been 
shown in Florida, which has intensive agricultural activity, humid climate, frequent heavy 
rainfall, and widespread use of irrigation and drainage. Several studies have shown the 
extent of P leaching that can occur in deep, sandy soils. 

In 1990, the Middle Suwannee River area was approved as a Hydrologic Unit 
Area project based on data generated by the Florida Department of Environmental 
Protection. These data showed an elevated concentration of nitrate-nitrogen in the 
Floridan Aquifer in the Suwannee River Basin, especially in areas of intensive 
agricultural activity. Phosphorus concentrations in the Suwannee River ranged from 0.40 
to 0.49 mg/L which were 6.4 times the median regional value of north Florida streams. 
The Hydrologic Unit Area program was developed to reduce or prevent water quality 
degradation of the Floridan Aquifer and the Suwannee River resulting from agricultural 
operations. Management of nutrients (potential contaminants) in dairy waste effluent 
through spray field crop production systems is an important component in the overall 
scheme for protecting ground and surface water from elevated levels of N and P. The use 
of inappropriate crop management technology under a dairy effluent irrigation system 
can lead to the loss of N to the ground water. Uptake of nutrients by agronomic crops 
sequenced over time is an effective, economical, and environmentally sound means of 



110 

nutrient recovery. Cropping systems designs are needed to meet environmental demands 
by maximizing nutrient uptake while meeting the needs of dairy producers. 

The Use of Dairy Manure Effluent in A Rhizoma (PerenniaH Peanut Based 
Cropping Systems for Nutrient Recovery and Water Quality Enhancement was a research 
project established under the Hydrologic Unit Area project (HUA). The objective of this 
project was to evaluate five cropping systems grown under a dairy effluent disposal 
irrigation system, comparing their effectiveness in nutrient recovery and maintenance of 
acceptable levels of N and P in ground water. The cropping systems were corn-forage 
sorghum-rye, com-bermuda grass-rye, bermuda grass-rye, perennial peanut-rye, and 
corn-perennial peanut-rye and the N application rates were (448,672 and 896 kg/ha/yr) 
which correspond to P loadings of (1 12, 168 and 224 kg/ha/yr). My study was a 
component of this project and addressed P forms and retention in the soil profile under 
two cropping systems (corn-forage sorghum-rye and perennial peanut-rye) and two N 
application rates (448 and 896 kg/ha/yr) which correspond to phosphorus loadings of 
(112 and 224 kg/ha/yr). The corn-forage sorghum-rye cropping system represents 
traditional crops for the Middle Suwannee River area and the perennial peanut-rye system 
is an improved cropping system recently introduced to the area. 

The main objective of this research was to study the effect of dairy waste effluent 
application on P accumulation, forms, and retention in the soil profile of a sandy soil 
under two cropping systems. The specific objectives and hypotheses of this research were 
as follows: ^ 

Objective 1: Quantify and characterize inorganic P forms in the soil profile of the 
chosen cropping systems with increasing effluent P application. 



Ill 

Hypothesis: Application of dairy effluent will increase P levels in the soil 
resulting in an accumulation of P in the soil profile. 

Objective 2: Quantify and characterize P retention in the soil profile. 

Hypothesis: Soil retention capacity will decrease with continuous addition of 
dairy effluent and may induce a downward movement of P. 

Objective 3: Determine P uptake by the chosen cropping systems under two rates 
of effluent appHcation. ■•' ■ 

Hypothesis: P accumulation in soil profile will decrease with increasing plant 
uptake. 

The effect of effluent application on P accumulation was discussed in Chapter 2. 
Phosphorus forms and fractionation data are presented in Chapter 3 while P retention was 
discussed in Chapter 4. Downward P movement was assessed in Chapter 5 and P removal 
by the cropping systems was presented in Chapter 6. In this chapter, the most important 
finding of this research are summarized according to the objectives and hypotheses 
mentioned above. In addition, future research topics are identified. 

Investigation of the P levels at the study site indicated that the soil appears to have 
been heavily loaded with animal waste prior to the start of this study, although amounts 
and dates could not be established. The application of dairy effluent during the study 
period resulted in increased TP level through out the soil profile. Soil test P (Mehlich I) 
increased for the P-R cropping system under the high application rate. Soil test P 
(Mehlich I) was highly correlated with Ca, Mg, Al and Fe extracted by Mehlich I 
solution, with 93% of the variability explained by this relationship. 



112 

Al- and Fe-associated P constituted the major proportion (62%) of the TP in the 
soil profile. Labile-P constituted 18 to 40% of TP throughout the profile with an 
increasing trend at the lower depth of the soil profile by the end of the study period. 
Labile P is defined as easily removable P and its increase in the lower depths is a clear 
indication of a vertical P movement. In fact, the application of dairy effluent during the 
study period increased all P pools significantly throughout the soil profile except Ca- and 
Mg-associated P which remained constant. 

The ability of the soil at the study site to retain P was low in comparison to soil 
from the native site. The relative P adsorption (RPA) for the study site soil was 0.5 to 0.6 
while the native soil RPA was 0.8 to 0.9. The lower retention of P for this soil was 
associated with about 50% P saturation at the surface soil. The Degree of P Saturation 
(DPS) for the surface soil from the native site was about 17%. Higher DPS values for the 
surface soil suggests that the surface horizon is more likely to release P than the deeper 
soil depths. Degree of P Saturation showed a strong relation with EPCo and Po which 
suggests that DPS can be used in the assessment of the tendency of this soil to release P. 
Downward P movement in the soil profile suggested by water soluble phosphorus (WSP) 
and labile P data. Both parameters, during the two years of effluent application showed a 
significant increase in the lower soil depths. '^'" -') ■ ' ^> '^ 

Phosphorus removal by the cropping systems was higher for the C-FS-R than the 
P-R cropping system. However, P uptake by the cropping systems did not reduce the high 
level of soil P that was already present before effluent application. This study suggests 
that when STP levels in the soil exceed optimum values for crop production, the 



,113 - \ ' . ■ 

application of dairy waste based on estimated crop N requirements may not be 
appropriate on heavily P loaded sandy soil such as at the study site. 

Further research is needed in the area of linking traditional soil test P with 
environmentally oriented P tests such as WSP. The results of the significant linear 
relationship between soil test P (Mehlich I), labile P and WSP, and labile P and WSP and 
DPS-1 and DPS-2 under the conditions of this study could be useful for future 
comparison with similar results of other soils. Also, DPS is another area for future 
research. Values of DPS - 1, calculated from oxalate extractable-P and DPS - 2, 
calculated from Mehlich I for soil samples from the study site were highly correlated (r^ 
- 0.92) which suggest the possibility of integrating the DPS to the most common soil test 
P in the region. Other correlations, such as equilibrium P concentration (EPCo) and DPS - 
1 (r^ = 0.94), and soluble P (Po) and DPS - 1 (r^ =0.92) suggest that an integration of such 
tools could be used in the study of the assessment of the tendency of this soil to release P. 



APPENDIX 
SELECTION OF SOIL: SOLUTION RATIO 

The solid: solution ratio is critical in sequential extraction. The use of excess solid 
can result in incomplete dissolution of target phases due to saturation of the solution with 
respect to the target phase (Ruttenberg 1990). A modification for solid: solution is always 
needed for a specific sample (Ruttenberg 1992). Therefore, prior to the selection of the 
soil: solution ratio used in this study, three soil samples from the study site were used for 
comparison with three different soil: solution ratios. The soil samples (Ic, 3c, and 6c) 
were from three different depths; 0-15, 30-45, and 80-100 cm, respectively. The soil: 
solution ratio tried were; 1/10, 1/25, and 1/50. All samples were subjected to a complete 
fractionation scheme, as explained below, with two modified procedures. The first, 
include the complete procedure with three 1 MNH4CI (adjusted to pH 7.0) extractions, 
and the second include the complete procedure with one 1 MNH4CI (adjusted to pH 7.0) 
extractions. 

In a third trial, the same soil samples were subjected to a complete fractionation 
scheme with two soil; solution ratios (1/50 and 1/100) and three 1 A/NILtCl (adjusted to 
pH 7.0) extractions. Total P (TP) for the three samples used in these trials were determine 
by ashing 1.0 g of soil for 3 hours and then solubilizing with 6 M HCl (Anderson, 1976). 
Their TP were 816, 304, and 111, respectively. 

The preliminary data of the two trials were shown in the tables (3-1, 3-2, and 3-3) 
below. A careftil examination of the data in term of P concentration in each trial and 



114 



115 

extraction suggested that the 1/25 soil: solution ratio and two 1 MNH4CI (adjusted to pH 
7.0) extractions seem to be appropriate for this soil. 
Table 1-1. Data of trial (1), soil: solution ratio selection 



Sample 


Ratio 


NH4CII 


NH4CI2 


NH4CI3 

- P(mg/kg) 


NaOH 


HCl 








Ic 


1/10 


14 


16 


16 


551 


277 


Ic 


1/10 


14 


16 


16 


581 


272 



Ic 1/25 24 13 17 654 93 
Ic 1/15 24 24 17 804 90 



Ic 


1/50 


21 


22 


18 


566 


76 


Ic 


1/50 


23 


18 


10 


561 


74 


3c 


1/10 


6 


4 


9 


219 


138 


3c 


1/10 


6 


4 


4 


216 


133 


3c 


1/25 


9 


1 


5 


353 


29 


3c 


1/15 


7 


2 


1 


338 


36 


3c 


1/50 


4 





1 


323 


38 


3c 


1/50 


4 








309 


34 


6c 


1/10 








. 


88 


17 


6c 


1/10 











84 


18 


6c 


1/25 











86 


22 


6c 


1/15 











94 


22 


6c 


1/50 











105 


32 


6c 


1/50 











97 


31 



116 



Table 1-2. Data of trial (2), soil: solution ratio selection 
Sample Ratio NH4CI 1 NaOH HCl 

P (mg/kg) 



Ic 1/10 15 207 584 
Ic 1/10 15 566 241 



Ic 


1/25 


27 


599 


325 


Ic 


1/15 


26 


746 


260 


Ic 


1/50 


22 


596 


97 


Ic 


1/50 


23 


659 


1Q2 


3c 


1/10 


7 


333 


71 


3c 


1/10 


7 


319 


90 


3c 


1/25 


10 


380 


41 


3c 


1/15 


3 


384 


40 


3c 


1/50 





356 


39 


3c 


1/50 





385 


36 


6c 


1/10 





63 


29 


6c 


1/10 





93 


16 


6c 


1/25 





89 


22 


6c 


1/15 





88 


21 


6c 


1/50 





115 


33 


6c 


1/50 





111 


30 



117 



Table 1-3. Data of trial (3), soil: solution ratio selection 

Sample Ratio NH4CI 1 NH4CI2 NH4CI3 NaOH HCl 

P(mg/kg) 



Ic 1/50 49 39 28 

Ic 49 36 26 

3c 31 20 14 

3c 29 19 15 

6c 444 

6c 4 4 ' 4 

Ic 1/100 70 51 27 

Ic 77 49 33 

3c 42 29 19 

3c 37 24 19 

6c 7 7 6 

6c 7 7 6 



610 


47 


602 


44 


432 


23 


388 


19 


126 


12 


118 


12 


695 


54 


374 


218 


415 


21 


398 


20 


145 


12 


143 


13 



LIST OF REFERENCES 

American Public Health Association (APHA). 1985. Standard Methods for the 

Examination of water and Wastewater, 16"' ed., Washington, DC: American 
Public Health Association. 

Anderson, J.M. 1976. An ignition method for determination of total phosphorus in lake 
sediments. Water Res. 10:329-331. 

Bache, B. W., & E.G. Williams. 1971. A phosphate sorption index for soils. J. Soil Sci. 
22:289-301. 

Bamhisel, R., & P.M. Bertsch. 1982. Aluminum. In A.L. Page, R.H. Miller, & D.R. 

Keeney (eds.), Methods of Soil Analysis Part 2. Chemical and Microbiological 
Properties, 2""* ed. Madison, WI: ASA/SSSA,. 

Beegle, Douglas, Andrew Sharpley, & Don Graetz. 1998. Interpreting soil test 

phosphorus for environmental purposes. In J. Thomas Sims (ed.). Soil Testmg for 
Phosphorus: Environmental Uses and Implication. SERA-IEG 17 USDA-CREES 
Regional Committee: Minimizing Agricultural Phosphorus Losses for protection 
of the Water Resources. Bulletin No. 389. 

Boggess, C.F., E.G. Flaig, & R.C. Fluck. 1995. Phosphorus budget-basin relationships for 
Lake Okeechobee tributary basin. Ecol. Eng. 5:143-162. 

Bowman, R.A., Rodriguez,J.B. and Self, J.R. 1998. Compersion of methods to estimate 
occluded and resistant soil phosphorus. Soil Sci. Soc. Am. J. 62:338-342. 

Brady, N.C. 1990. The Nature and Properties of Soil. 10* ed. Upper Saddle River, NJ: 
Prentice-Hall. 

Breeuwsma, A., J.G.A. Reijerink, & O.F. Schoumans.1995. Impact of manure on 

accumulation and leaching of phosphate in areas of intensive livestock farming. P. 
239-251. In K. Steele (ed.). Animal waste and the land-water interface. New 
York: Lewis PubUshers-CRC. 

Broadbent, F.E. 1965. Organic Matter. In Black, C.A ed.. Methods of Soil Analysis, Part 
2, pp. 539-580. 

Chang. S.C, & M.L. Jackson. 1957. Fractionation of soil phosphorus. Soil Sci. 84:133- 
144. 



118 



119 



Copee, J.T., C.E. Evans, & H.C. Williams. 1981. Soil test fertilizer recommendations for 
Alabama crops. Alabama Agric. Exp. Stn. Circ. 25 1 . 

Daniel,T.C., A.N. Sharpley, & J.L. Lemunyon. 1998. Agriculture phosphorus and 
eutrophication: A symposium overview. J. Environ. Qual. 27:251-257. 

Dantzman, C.L., M.F. Richter, and F.G. Martin. 1983. Chemical elements in soils under 
cattle pens. J. Environ. Qual. 12; 164- 168. 

Day, P.R. 1965. Pipette method of Particle Size Analysis. In C.A. Black (ed.). Methods 
of Soil Analysis, Part 1 Agronomy. 9: 552-562. Amer. Soc. of Agron, Madison, 
WI. 

Dooley, D.O. 1996. The status of phosphorus in natural wetland receiving dairy lagoon 
effluent. MS. Thesis. University of Florida, Gainesville. 

Dooley, D.O., V.D. Nair, & D.A. Graetz. 1993. Phosphorus release from beef and dairy 
manures and manure-contaminated soils. P.29. In agronomy abstracts. Madison, 
WI: ASA. 

Eghball, B., G.D. Binford, &. D.D. Baltensperger. 1996. Phosphorus movement and 
adsorption in a soil receiving long-term manure and fertilizer application. J. 
Environ. Qual. 25:1339-1343. 

Federico, A., K.G. Dickson, C.R. Kratzer, & F.E. Davis. 1981. Lake Okeechobee water 
quaUty studies and eutrophication assessment. Tech. Publ. 81-2. West Pahn 
Beach, FL: South Florida Water Management District. 

French, E.G., K.R. Woodard, D.A. Graetz, G.M.Prine, & H.H. Van Horn. 1995. A second 
year report on the use of dairy manure effluent in a rhizoma (perennial) peanut 
based cropping system for nutrient recovery and water quality enhancement. In 
Proc. 32nd Annual Florida Dairy Production Conference. 11-12 April 1 995 . 
Gainesville: University of Florida, Gainesville, FL. 

Graetz, D.A. 1989. Phosphorus levels in soils of selected Lake Okeechobee watershed 

dairies. P. 1 13-1 18. In Proc. 26* Annual Florida Dairy Production Conference. 2- 
3 May 1989. Gainesville: University of Florida. 

Graetz, D.A., & V.D. Nair. 1995. Fate of phosphorus m Florida spodosols contaminated 
with cattle manure. Ecol. Eng. 5 : 1 63- 1 8 1 . 

Graetz, D.A., & V.D. Nair. 1999. Inorganic forms of phosphorus in soils and sediments, 
pp 171-182. In K. R. Reddy, G. A. O'Connor, «& C.L.Schelske (eds.) Phosphorus 
Biogeochemistry in Subtropical Ecosystems. Boca Raton, FL: CRC Press. 



120 



Graetz, D.A., V.D. Nair, & K.R. Reddy. 1991. Biogeochemical behavior and transport of 
phosphorus in the Lake Okeechobee Basin. South Florida Water Management 
District, West Pahn Beach, FL. 

Harris, W.G., R.D. Rhue, G. Kidder, R.B. Brown, & R. Littell. 1996. Phosphorus 

retention as related to morphology of sandy coastal Plain materials. J. Environ. 
Qual. 60:1513-1521. 

Harris, W.G., H.D. Wang, &. K.R. Reddy. 1994. Dairy manure influence on soil and 
sediment composition: implication for phosphorus retention. J. Environ. Qual. 
23:1071-1081. 

Harris, Willie, &, Wade Hurt. 1999. Introduction to soils of subtropical Florida, pp 143- 
162. In K. R. Reddy, G. A. O'Connor, & C.L.Schelske (eds.), Phosphorus 
Biogeochemistry in Subtropical Ecosystems. Boca Raton, FL: CRC Press. 

Hieltjes, A.H.M., & L. Lijklema. 1980. Fractionation of inorganic phosphate in 
calcareous sediments. J. Environ. Qual. 9:405-407. 

King, L.D., J.C. Bums, & P.W. Westerman. 1990. Long-term swine lagoon effluent 

application on coastal bermuda grass: II. Effect on nutrient accumulation in soil. 
J. Environ. Qual. 19:756-760. 

Kingery, W.L., C.W. Wood, D.P. Delaney, J.C. Williams, & G.L. Mullins. 1994. Impact 
of long-term land application of broiler litter on environmentally related soil 
properties. J. Environ. Qual. 23:139-147. 

Lanyon, L.E. 1994. Dairy manure and plant nutrient management issues affecting water 
quality and the dairy industry. J. Dairy Sci. 77; 1999-2007. 

Lemunyon, J., & T.C. Daniel. 1998. Phosphorus management for water quality 
protection: A national effort. In J. Thomas Sims, (ed.). Soil Testing for 
Phosphorus: Environmental Uses and Implication. SERA-IEG 17 USDA-CREES 
Regional Committee: Minimizing Agricultural Phosphorus Losses for protection 
of the Water Resources. BuUetm No. 389. 

Manure Master. 1999. http://netral.ftw.nrcs.usda.gov/ManureMaster/MM22.html 

Mckeague, J.A., & J.H. Day. 1966. Dithhionite and oxalate-extractable Fe and Al as aids 
in differentiating various classes of soils. Can. J. Soil Sci. 46:13-22. 

Mehlich, A. 1953. Determination of p, Ca, Mg, K, Na and NH4. Soil Testing Div. Publ. 
1-53, Raleigh, NC: North Carolina Department of Agriculture. 

Menzies, N.W., j.A. Skilton, & C.N. Guppy. 1999. Phosphorus storage on effluent 
irrigated land. J. Environ. Qual. 28:750-754. 



121 



Moore, Phillip A. Jr., Brad C. Joem, & Tony L. Provin. 1998. Improvmrnt needed in 

environmental soil testing. In Sims, J. Thomas (ed.) Soil Testing for Phosphorus: 
Environmental Uses and Implication. SERA-IEG 17 USDA-CREES Regional 
Committee: Minimizing Agricultural Phosphorus Losses for protection of the 
Water Resources. Bulletin No. 389. 

Mozaffari, M., & J.T. Sims. 1994. Phosphorus availability and sorption in an Atlantic 

Coastal Plain watershed dominated by animal based agriculture. Soil Sci. 157:97- 
107. 

Mozaffari, M., & J.T. Sims. 1996. Phosphorus transformation in poultry litter-amended 
soils of the Atlantic Coastal Plain. J. Environ. Qual. 25:1357-1365. 

Murphy, J., & J.P. Riely. 1962. A modified single solution method for the determination 
of phosphate in natural waters. Anal. Chim. Acta 27:31-36. 

Nair, P.S., T.J. Logan, A.N. Sharpley, L.E. Sommers, M.A. Tabatabai, & T.L. Yuan. 
1984. Interlaboratory comparsion of a standardized phosphorus adsorption 
procedure. J. Environ. Qual. 13:591-595. 

Nair, V.D., D.A. Graetz, & K.M. Portier. 1995. Forms of phosphorus in soil profile from 
dairies of South Florida. J. Environ. Qual. 59:1244-1249. 

Nair, V.D., D.A. Graetz, 8c K.R. Reddy. 1998. Dairy manure influences on phosphorus 
retention capacity of Spodosols. J. Environ. Qual. 27:522-527. 

Nair, V.D., R.R. Villapando, & D.A. Graetz. 1999. Phosphorus retention capacity of 
Spodic horizon under varying environmental conditions. J. Environ. Qual. 
,.H 28:1308-1313. 

Olsen, S.R. & F.E. Khasawneh. 1980. Use and limitation of physical-chemical criteria for 
assessing the status of phosphorus in soils, pp. 361-410. In F.E. Khasawneh, E.C. 
Sample & E.J. Kamprath (eds.). The Role of Phosphorus in Agriculture. Madison, 
WI: SSSA. 

Olsen, S.R., & L.E. Sommers. 1982. Phosphorus. In A.L. Page, R.H. Miller & D.R. 

Keeney (eds.). Methods of Soil Analysis Part 2. Chemical and Microbiological 
Properties, T^ ed. Madison, WI: ASA/SSSA. 

Parry, R. 1998. Agricultural phosphorus and water quality: A U. S. Environmental 
Protection Agency perspective. J. Environ. Qual. 27:258-261. 

Pettersson, K., & V. Istvanovics. 1988. Sediment phosphorus in Lake Balaton: Forms and 
Mobility. Ergeb. Limnol. 30:25-41. 



122 



Pote, D.H., T.C. Daniel, A.N. Sharpley, P.A. Moore, Jr., D.R. Edwards, & D.J. Nichols. 
1996. Relating extractable soil phosphorus to phosphorus losses in runoff. Soil 
Sci. Soc. Am. J. 60:855-859. 

Project C-2: Use of Dairy Manure Effluent in a Rhizoma (Perennial) Peanut Based 

Cropping System for nutrient Recovery and Water Quality Enhancement. 1995, 
Gainesville:IFAS, University of Florida. 

Reddy, K.R., E.G. Flaig, & D.A. Graetz. 1996. Phosphorus storage capacity of uplands, 
wetlands and streams of the Lake Okeechobee Watershed, Florida. Agriculture, 
Ecosystems and Environment 59: 203-216. 

Reddy. K.R., E. Lowe, & T. Fotaine. 1999. Phosphorus in Florida's ecosystems: analysis 
of current issues, pp 1 1 1-138. In K. R. Reddy, G. A. O'Connor, & C.L.Schelske 
(eds.) Phosphorus Biogeochemistry in Subtropical Ecosystems. Boca Raton, FL: 
CRC Press. 

Reddy, K.R., G.A. O'Connor, & P.M. Gale. 1998. Phosphorus sorption capacities of 

wetland soils and sediments impacted by dairy effluent. J. Environ. Qual. 27:438- 
447. 

Rhue, R.D. & Willie G. Harris. 1999. Phosphorus sorption/desorption reactions in soils 
and sediments, pp 187-203. In K. R. Reddy, G. A. O'Connor, & C. L. Schelske 
(eds.). Phosphorus Biogeochemistry in Subtropical Ecosystems. Boca Raton, FL: 
CRC Press. 

Rhue, R.D, W.G. Harris, G. Kidder, R. B. Brown, and R. C. Brown. 1994. A soil-based 
phosphorus retention index for animal waste disposal on sandy soil. Final Project 
Report. Florida Department of Environmental Protection, Tallahassee, FL. 

Ruttenberg, Kathleen C. 1990. Diaggenesis and burial of phosphorus in marine 

sedunents: Implication for the marine phosphorus budget. Ph.D. thesis, Yale 
Univ. 375p. 

Ruttenberg, Kathleen C. 1992. Development of a sequential extraction method for 
different forms of phosphorus in marine sediments. Limnol. Oceanogr. 
37(7): 1460-1482. 

Ryden, J.C, &. J.K. Syers. 1977. Origin of the labile phosphate pool m soils. Soil Sci. 
123:353-361. 

SAS Institute Inc. 1985. SAS User's Guide: Statistics, Version 5th. Ed. Cary, NC: SAS. 

Schabenberger, Oliver. 1996. A Brief Comparison of PROC GLM and PROC MIXED of 
SAS. http://www.cas.vt.edu/schabenb/GLMvsMIXED.html 



123 



Sharpley, A.N., Chapra, S.C, Wedepohl, R., Sims, J.T., Daniel, T.C. & Reddy, K.R. 

1994b. Managing agricultural phosphorus for protection of surface waters: issues 
and options. J. Environ. Qual. 23:437-451. 

Sharpley, A.N., & A.D. Halvorson. 1994. The management of soil phosphorus 

availability and its impact on surface water quality, pp 8-72. In R.Lal & B.A. 
Stewart (eds.) Soil Processes and Water Quality. CRC Press, Boca Raton, FL. 

Sharpley, A.N., & S. Rekolainen. 1997. Phosphorus in agriculture and its environmental 
implication. In Tunney, H., O.T. Carton, P.C. Brookes, & A.E. Johnston (eds.) 
Phosphorus Loss from Soil to Water. CAB International, New York. 

Sharpley, A. N., S.J. Smith, B.A. Stewart, & A.C. Mathers. 1984. Forms of phosphorus in 
soils receiving cattle feedlot waste. J. Environ. Qual. 13:21 1-215. 

Sharpley, A.N., & J.K. Syers. 1979. Phosphorus inputs into a stream draining an 

agricultural watershed: II. Amounts and relative significance of runoff types. 
Water, Air and Soil Pollut. 1 1 :41 7-428. 

Sharpley, Andrew, T.C. Daniel, J.T. Suns, & D.H. Pote. 1996. Determining 

environmentally sound soil phosphorus levels. J. Soil and Water Cons. 5 1(2): 160- 
166. 

Sibbesen, E., & A.N. Sharpley. 1997. Setting and justifying upper critical limits for 

phosphorus in soils. In Tunney, H., O.T. Carton, P.C. Brookes, & A.E. Johnston 
(eds.) Phosphorus Loss from Soil to Water. CAB International, New York. 

Simard, R.R., D. Vluis, G. Gangbazo, & S. Beauchemin. 1995. Phophorus status of forest 
and agricultural soils from a watershed of high anunal density. J. Environ. Qual. 
24:1010-1017. 

Sims, J.T., & K.L. Gartley. 1996 Nutrient management handbook for Delaware. Coop. 
Bult. 59. Univ. of Delaware, Newark, DE. 

Sims, J.T., R.R. Simard, & B.C. Joren. 1998. Phosphorus loss in agricultural drainage: 
Historical perspective and current research. J. Environ. Qual. 27:227-293. 

Sims, J. Thomas, Steve Hodges, & Jessica Davis. 1998. Soil testing for phosphorus: 

current status and uses in nutrient management program. In Sims, J. Thomas (ed.) 
Soil Testing for Phosphorus: Environmental Uses and Implication. SERA-IEG 1 7 
USDA-CREES Regional Committee: Minimizing Agricultural Phosphorus Losses 
for protection of the Water Resources. Bulletin No. 389. 

Sjoerd, E.A., T.M. van der Zee, & Willem H. Van Riemsduk. 1988. Model for long-term 
phosphate reaction kinetics in soil. J. Environ. Qual. 17:35-41. 



i,j.JHIlllill»J!Piira!Wai"W*»l!'-'^ 



124 



Soil Survey Staff. 1999. Soil Taxonomy: A basic system for soil classification for making 
and interpreting soil surveys, 1999. Agriculture hand book 436.U.S. Government 
Printing Office, Washington, DC. 

Staples, R. Charles, Robert S. Tervola, & Edwin C. French. 1997. Forage Production 
Practices by Dairy Producers in the Suwannee Valley. Gainesville: Cooperative 
Extension Service, Institute of Food and Agriculture Science, Univ. of Florida. 
Circular 1049. 

Sweeten, J.M., M.L. Wolfe, E.S. Chasteen, M. Sanderson, B.A. Auvermann, & G.D. 
Alston. 1995. Dairy lagoon effluent irrigation: Effect on runoff quality, soil 
chemistry, and forage yield. P. 99-106. In K. Steele(ed.) Animal waste and the 
land-water interface. Lewis Publishers-CRC, New York. 

Thomann, R.V., & J.A. Mueller. 1987. Principles of surface water quality modeling and 
control. Harper Collins Publ., New York. 

Tiessen, H., and J.O. Moir. 1993. Characterization of available P by sequential extraction 
p.75-86. In M.R. Carter (ed.) Soil sampling and method of analysis. Can. Soc. 
Soil Sci. Lewis Publishers.BocRaton, FL. 

Tisdale, Samuel L., Werner L. Nelson, James D. Beaton, & John L. Havlin. 1993. Soil 
Fertility and Fertilizers. 5* ed. Macmillan Publishing Company, New York. 

U.S. Environmental Protection agency (USEPA). 1996. Environmental indicators of 
water quality m the United states. EPA 841-R-96-002. USEPA Office of Water 
(4503F), U.S. Government Printing Office, Washington, DC. 

U.S. Environmental Protection Agency, 1988. Nonpoint source pollution in the US: 
report to Congress. Office of Water, Criteria and Standards division, USEPA, 
Washington, DC. 

Van Der Zee, S.E.A.T.M., & W.H. van Riemsdijk. 1991. Model for the reaction kinetics 
of phosphate with oxides and soils. In G.H. Bolt, M.F.DeBoodt, M.H.B. Hayes, & 
M.B. McBride (eds.). Interaction at the Soil Colloid-Soil solution Interfece. 
Kluwer Academic Publ., Dordrecht, pp.205-239. 

Van Der Zee, S.E.A.T.M., L.G.J. Fokkink, & W.H. van Riemsdijk. 1987. A new 

technique for assessment of reversibly adsorbed phosphate. Soil Sci. Soc. Am. J. 
51:599-604. 

Van Horn , H.H., A.C. Wilkie, W.J. Powers, & R.A. Nordstedt.l994.Components of 
dairy manure management systems. J. Dairy Sci. 77:2008-2030. 



125 



Van Horn, H.H., G.L. Newton, R.A. Nordstedt, E.G. French, G. Kidder, D.A. Graetz, & 
C.F. Chambliss. 1998. Dairy manure management: strategies for recycling 
nutrients to recover fertilizer value and avoid environmental pollution. 
Cooperative Extension Service, Institute of Food and Agriculture Science, Univ. 
of Florida. Circular 1016. 

Wang, H.D., W.G. Harris, & K.R. Reddy. 1992. Stability of P forms in dairy-influenced 
soils under leaching conditions. P.63. In Agronomy abstract. ASA, Madison, WI. 

Wang, H.D., W.G. Harris & K.R. Reddy. 1995. Stability of? forms in dairy-impacted 
soils under simulated leaching. Ecol. Eng. 5:209-227. 

Williams, J.D.H., J.K. Syers, & T.W. Walker. 1967. Fractionation of soil inorganic 
phosphate by a modification of Chang and Jackson's procedure. Soil Sci. Soc. 
Am. Proc. 31:736-739. 

Wolfmger, R. D. and M.Chang, 1996. Comparing the SAS GLM and MIXED procedure 
for Repeated Measures. SAS institute Inc., Cary, NC. 

Woodard, K. R., E.C. Fremch, D.A. Graetz, L.E. SoUenberger, L.A. Sweat, B.L. Wade, 
S.J. Rymph, B. Macoon, G.M. Prine, and H.H. Van Horn. 2000. Utilzation of 
dairy waste effluent in year-round cropping systems for nutrient recovery and 
water quality enhancement. (In press). 

Yuan, T.L., & Dawn E. Lucas. 1982. Retention of phosphorus by sandy soils as evaluated 
by adsorption isotherms. Florida Agr. Expt. Stas. Jour. Ser. No. 3136. 

Zhang, T.Q., & A.F. Mackenzie. 1997. Changes in soil phosphorus fractions under long- 
term com monoculture. Soil Sci, Am. J. 61:485-493. 



'l' :'■ 



BIOGRAPHICAL SKETCH 

Abdullah Alshankiti was bom on February 13, 1956, in Alqunfodah, Saudi 
Arabia. He completed his elementary school in his hometown of Alqunfodah. He moved 
to Riyadh, Saudi Arabia, to finish high school, and in 1979 he received a bachelor of 
science degree in agriculture from Riyadh University. In the same year, he joined the 
Department of Soil and Irrigation at the National Agriculture and Water Research Center, 
Riyadh, as a research assistant. In 1989, he obtained a Master of Science in Agriculture 
degree from California State University, Chico, and returned to work at the National 
Agriculture and Water Research Center, Riyadh. After earning his doctoral degree in soil 
and water science from the University of Florida in May 2000, he will return to his job at 
the National Agriculture and Water Research Center, Riyadh, Saudi Arabia. 



126 



I certify that I have read this study and that in my opinion it conforms to 
acceptable standards of scholarly presentation and is fully adequate, in scope and quality, 
as a dissertation for the degree of Doctor of Philosophy. 



D,.Jtd 




Donald A. Graetz, Chair 
Professor of Soil and Water Science 



I certify that I have read this study and that in my opinion it conforms to 
acceptable standards of scholarly presentation and is fully adequate, in scope and quality, 
as a dissertation for the degree of Doctor of Philosophy. 




R 

Professor of Soil and Water Science 

I certify that I have read this study and that in my opinion it conforms to 
acceptable standards of scholarly presentation and is fully adequate, in scope and quality, 
as a dissertation for the degree of Doctor of Philosophy. 

Willie G. Harris, Jr. ^~^ 

Professor of Soil and Water Science 

I certify that I have read this study and that in my opinion it conforms to 
acceptable standards of scholariy presentation and is fully adequate, in scope and quality, 
as a dissertation for the degree of Doctor of Philosophy. 




..^^..^^^-^ 



Roger A. Nordstedt 
Professor of Agricultural and Biological 
Engineering 
This dissertation was submitted to the Graduate Faculty of the College of 
Agricultural and Life Sciences and to the Graduate School and was accepted as partial 
fulfillment of the requirements for the degree of Doctor of Philosophy. 



May 2000 



Dean, College of AgricultHraland Life 
Sciences 



Dean, Graduate School 






\20C/> 
,H6l 



UNIVERSITY OF FLORIDA 



3 1262 08555 1660